Neuropathic pain The pharmacological management of neuropathic pain in adults in non-specialist settings

NICE clinical guideline Draft for consultation, September 2011

This guideline was developed following the NICE short clinical guideline process. This document includes all the recommendations, details of how they were developed and summaries of the evidence they were based on.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 1 of 150 Contents

Introduction ...... 3 Drug recommendations ...... 5 Who this guideline is for ...... 6 Patient-centred care ...... 7 1 Recommendations ...... 8 The following definitions apply to this guideline...... 8 List of all recommendations ...... 8 2 Care pathway ...... 15 3 Evidence review and recommendations ...... 18 3.1 Methodology ...... 18 3.2 Painful diabetic neuropathy (PDN) ...... 30 3.3 Post-herpetic neuralgia (PHN) ...... 65 3.4 Other neuropathic pain conditions ...... 85 4 Notes on the scope of the guideline ...... 131 5 Implementation ...... 131 6 Other versions of this guideline ...... 131 6.1 Quick reference guide ...... 131 6.2 NICE pathway ...... 131 6.3 ‘Understanding NICE guidance’ ...... 132 7 Related NICE guidance ...... 132 8 Updating the guideline ...... 132 9 References ...... 133 10 Glossary and abbreviations ...... 143 Appendix A Contributors and declarations of interests ...... 146

Appendices 10.1 to 10.11 are in separate files.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 2 of 150

This clinical guideline updates and replaces the following recommendations on the drug treatment of painful diabetic neuropathy in previous NICE clinical guidelines:

recommendations 1.11.5.2, 1.11.5.3, 1.11.5.4, 1.11.5.5 and 1.11.5.7 in ‘Type 1 diabetes: diagnosis and management of type 1 diabetes in children, young people and adults’ (NICE clinical guideline 15) recommendations 1.14.2.3, 1.14.2.4, 1.14.2.5 and 1.14.2.6 in ‘Type 2 diabetes: the management of type 2 diabetes’ (NICE clinical guideline 87).

Introduction

Pain is an unpleasant sensory and emotional experience that can have a significant impact on a person’s quality of life, general health, psychological health, and social and economic well-being. The International Association for the Study of Pain (IASP 2011) defines: neuropathic pain as ‘pain caused by a lesion or disease of the somatosensory nervous system’. This is further delineated as central neuropathic pain ‘pain caused by a lesion or disease of the central somatosensory nervous system’, and peripheral neuropathic pain ‘pain caused by a lesion or disease of the peripheral somatosensory nervous system’.

Neuropathic pain is very challenging to manage because of the heterogeneity of its aetiologies, symptoms and underlying mechanisms (Beniczky et al. 2005). Examples of common conditions that have peripheral neuropathic pain as a symptom are painful diabetic neuropathy, post-herpetic neuralgia, trigeminal neuralgia, radicular pain, pain after surgery and neuropathic cancer pain (that is, chemotherapy-induced neuropathy and neuropathy secondary to tumour infiltration). Examples of conditions that can cause central neuropathic pain include stroke, spinal cord injury and multiple sclerosis. Neuropathic pain can be intermittent or constant, and spontaneous or provoked. Typical descriptions of the pain include terms such as shooting, stabbing, like an

Neuropathic pain: NICE clinical guideline DRAFT September 2011 3 of 150 electric shock, burning, tingling, tight, numb, prickling, itching and a sensation of pins and needles. People may also describe symptoms of allodynia (pain caused by a stimulus that does not normally provoke pain) and hyperalgesia (an increased response to a stimulus that is normally painful) (McCarberg 2006).

A review of the epidemiology of chronic pain found that there is still no accurate estimate available for the population prevalence of neuropathic pain (Smith and Torrance 2010). For example, the prevalence of neuropathic pain overall has been estimated at between 1% and 2%, based on summed estimates of the prevalence in the USA (Bennett 1997) and the UK (Bowsher et al. 1991). These estimates of population prevalence came from a number of heterogeneous studies of variable validity, are likely to be inaccurate and are inconsistent. Other condition-specific studies have also mirrored the heterogeneous nature of neuropathic pain. For example, painful diabetic neuropathy is estimated to affect between 16% and 26% of people with diabetes (Jensen et al. 2006; Ziegler 2008). Prevalence estimates for post- herpetic neuralgia range from 8% to 19% of people with herpes zoster when defined as pain at 1 month after rash onset, and 8% when defined as pain at 3 months after rash onset (Schmader 2002). The development of chronic pain after surgery is also fairly common, with estimates of prevalence ranging from 10% to 50% after many common operations (Shipton 2008). This pain is severe in between 2% and 10% of this subgroup of patients, and many of the clinical features closely resemble those of neuropathic pain (Jung et al. 2004; Mikkelsen et al. 2004; Kehlet et al. 2006). Furthermore, a study of 362,693 computerised records in primary care from the Netherlands estimated the annual incidence of neuropathic pain in the general population to be almost 1% (Dieleman et al. 2008). This considerable variability in estimates of the prevalence and incidence of neuropathic pain and similar conditions from general population studies is likely to be because of differences in the definitions of neuropathic pain, methods of assessment and patient selection (Smith and Torrance 2010).

Neuropathic pain: NICE clinical guideline DRAFT September 2011 4 of 150 Currently, a number of pharmacological treatments are commonly used in the UK to manage neuropathic pain in non-specialist settings. However, there is considerable variation in practice in terms of how treatment is initiated, whether therapeutic doses are achieved and whether there is correct sequencing of therapeutic classes. This may lead to inadequate pain control, with considerable morbidity. In the context of this guideline, non-specialist settings are defined as primary and secondary care services that do not provide specialist pain services. These include general practice, general community care and hospital care. Commonly used pharmacological treatments include antidepressants (tricyclic antidepressants [TCAs], selective serotonin reuptake inhibitors [SSRIs] and serotonin–norepinephrine reuptake inhibitors [SNRIs]), anti-epileptic (anticonvulsant) drugs (such as gabapentin, pregabalin and carbamazepine), topical treatments (such as capsaicin and lidocaine) and opioid analgesics. All of these drug classes are associated with disadvantages, as well as potential benefits. A further issue is that a number of commonly used treatments (such as amitriptyline) are unlicensed for treatment of neuropathic pain, which may limit their use by practitioners. There is also uncertainty about which drugs should be used initially (first-line treatment) for neuropathic pain, and the order (sequence) in which the drugs should be used.

This short clinical guideline aims to improve the care of adults with neuropathic pain by making evidence-based recommendations on the pharmacological management of neuropathic pain in non-specialist settings. A further aim is to ensure that those people who require specialist assessment and interventions are referred appropriately and in a timely fashion to a specialist pain service and/or other condition-specific services.

Drug recommendations

For all drugs, recommendations are based on evidence of clinical and cost effectiveness and reflect whether their use for the management of neuropathic pain is a good use of NHS resources. This guideline should be used in conjunction with clinical judgement and decision-making appropriate for the individual patient.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 5 of 150 The guideline will assume that prescribers will use a drug’s summary of product characteristics (SPC) and the British National Formulary (BNF) to inform decisions made with individual patients (this includes obtaining information on special warnings, precautions for use, contraindications and adverse effects of pharmacological treatments). However, the Guideline Development Group (GDG) agreed that having clear statements on drug dosage and titration in the actual recommendations is crucial for treatment in non-specialist settings, to emphasise the importance of titration to achieve maximum benefit.

This guideline recommends some drugs for indications for which they do not have a UK marketing authorisation at the date of publication, if there is good evidence to support that use. Where recommendations have been made for the use of drugs outside their licensed indications (‘off-label use’), these drugs are marked with a footnote in the recommendations. Licensed indications are listed in table 1.

Table 1 Licensed indications for recommended pharmacological treatments for neuropathic pain (August 2011) Amitriptyline Not licensed for neuropathic pain Duloxetine Licensed for painful diabetic neuropathy Gabapentin Licensed for peripheral neuropathic pain Imipramine Not licensed for neuropathic pain Lidocaine (topical) Licensed for post-herpetic neuralgia Nortriptyline Not licensed for neuropathic pain Pregabalin Licensed for central and peripheral neuropathic pain Tramadol Licensed for moderate and severe pain

Who this guideline is for

This document is intended to be relevant to healthcare professionals in non- specialist primary and secondary care settings. The target population is adults with neuropathic pain conditions. However, the guideline does not cover adults with neuropathic pain conditions who are treated in specialist pain services, or adults who have neuropathic pain in the first 3 months after trauma or orthopaedic surgical procedures.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 6 of 150 Patient-centred care

This guideline offers best practice advice on the pharmacological management of neuropathic pain in adults in non-specialist settings. Treatment and care should take into account patients’ needs and preferences. People with neuropathic pain should have the opportunity to make informed decisions about their care and treatment, in partnership with their healthcare professionals. If patients do not have the capacity to make decisions, healthcare professionals should follow the Department of Health’s advice on consent (available from www.dh.gov.uk/en/DH_103643) and the code of practice that accompanies the Mental Capacity Act (available from www.dh.gov.uk/en/SocialCare/Deliveringsocialcare/MentalCapacity). In Wales, healthcare professionals should follow advice on consent from the Welsh Government (available from www.wales.nhs.uk/consent).

Good communication between healthcare professionals and patients is essential. It should be supported by evidence-based written information tailored to the patient’s needs. Treatment and care, and the information patients are given about it, should be culturally appropriate. It should also be accessible to people with additional needs such as physical, sensory or learning disabilities, and to people who do not speak or read English.

If the patient agrees, families and carers should have the opportunity to be involved in decisions about treatment and care.

Families and carers should also be given the information and support they need.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 7 of 150 1 Recommendations

The recommendations in this clinical guideline are for the pharmacological management of neuropathic pain in non-specialist settings only. The Guideline Development Group acknowledged that there are other pharmacological and non-pharmacological treatments that will be of benefit to people with neuropathic pain, within different care pathways in different settings. However, the purpose of this clinical guideline is to provide useful and practical recommendations on pharmacological management in non- specialist settings for both people with neuropathic pain and healthcare professionals.

The following definitions apply to this guideline.

Non-specialist settings Primary and secondary care services that do not provide specialist pain services. Non-specialist settings include general practice, general community care and hospital care.

Specialist pain services Services that provide comprehensive assessment and multi-modal management of all types of pain, including neuropathic pain.

List of all recommendations

Key principles of care 1.1.1 Consider referring the person to a specialist pain service and/or a condition-specific service1 at any stage, including at initial presentation and at the regular clinical reviews (see recommendation 1.1.9), if:

they have severe pain or their pain significantly limits their daily activities and participation2 or

1 A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. 2 The World Health Organization ICF (International Classification of Functioning, Disability and Health) (2001) defines participation as ‘A person’s involvement in a life situation.’ It includes

Neuropathic pain: NICE clinical guideline DRAFT September 2011 8 of 150 their underlying health condition has deteriorated.

1.1.2 Continue existing treatments for people whose neuropathic pain is already effectively managed3.

1.1.3 Address the person’s concerns and expectations when agreeing which treatments to use by discussing:

the benefits and possible adverse effects of each pharmacological treatment why a particular pharmacological treatment is being offered coping strategies for pain and for possible adverse effects of treatment that non-pharmacological treatments are also available in non- specialist settings and/or through referral to specialist services (for example, surgical treatments and psychological therapies).

1.1.4 When selecting pharmacological treatments, take into account:

the person’s vulnerability to specific adverse effects because of comorbidities safety considerations and contraindications as detailed in the SPC patient preference lifestyle factors (such as occupation) any mental health problems (such as depression and/or anxiety4) any other medication the person is taking.

the following domains: learning and applying knowledge, general tasks and demands, mobility, self-care, domestic life, interpersonal interactions and relationships, major life areas, community, and social and civil life. 3 Note that there is currently no good-quality evidence on which to base specific recommendations for treating trigeminal neuralgia. The GDG expected that current routine practice will continue until new evidence is available (see also section 3.1). 4 Refer if necessary to ‘Anxiety’ (NICE clinical guideline 113), ‘Depression’ (NICE clinical guideline 90) and/or ‘Depression in adults with a chronic physical health problem’ (NICE clinical guideline 91) (available at www.nice.org.uk).

Neuropathic pain: NICE clinical guideline DRAFT September 2011 9 of 150 1.1.5 Explain both the importance of dosage titration and the titration process, providing written information if possible.

1.1.6 When withdrawing or switching treatment, taper the withdrawal regimen to take account of dosage and any discontinuation symptoms.

1.1.7 When introducing a new treatment, consider overlap with the old treatments to avoid deterioration in pain control.

1.1.8 After starting or changing a treatment, perform an early clinical review of dosage titration, tolerability and adverse effects to assess the suitability of the chosen treatment.

1.1.9 Perform regular clinical reviews to assess and monitor the effectiveness of the chosen treatment. Each review should include assessment of:

pain reduction adverse effects daily activities and participation5 (such as ability to work and drive) mood (in particular, whether the person may have depression and/or anxiety6) quality of sleep overall improvement as reported by the person.

5 The World Health Organization ICF (International Classification of Functioning, Disability and Health) (2001) defines participation as ‘A person’s involvement in a life situation.’ It includes the following domains: learning and applying knowledge, general tasks and demands, mobility, self-care, domestic life, interpersonal interactions and relationships, major life areas, community, and social and civil life. 6 Refer if necessary to ‘Anxiety’ (NICE clinical guideline 113), ‘Depression’ (NICE clinical guideline 90) and/or ‘Depression in adults with a chronic physical health problem’ (NICE clinical guideline 91) (available at www.nice.org.uk).

Neuropathic pain: NICE clinical guideline DRAFT September 2011 10 of 150 First-line treatment 1.1.10 Offer oral amitriptyline* or gabapentin as first-line treatment (see recommendation 1.1.11 for people with painful diabetic neuropathy). (For dosages please see box 1 Drug dosages).

1.1.11 For people with painful diabetic neuropathy, offer oral duloxetine as first-line treatment. If duloxetine is contraindicated, offer oral amitriptyline*. (For dosages please see box 1 Drug dosages).

1.1.12 Based on both the early and regular clinical reviews:

If improvement is satisfactory, continue the treatment; consider gradually reducing the dose over time if improvement is sustained. If amitriptyline* results in satisfactory pain reduction as first-line treatment but the person cannot tolerate the adverse effects, consider oral imipramine* or nortriptyline* as an alternative. If gabapentin results in satisfactory pain reduction as first-line treatment but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects, consider oral pregabalin as an alternative. (For dosages please see box 1 Drug dosages).

Second-line treatment 1.1.13 If satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person (see recommendation 1.1.16 for people with painful diabetic neuropathy):

If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with oral gabapentin (or

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 11 of 150 pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with gabapentin (or pregabalin) switch to or combine with oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects).

1.1.14 For people with painful diabetic neuropathy, if satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person:

If first-line treatment was with duloxetine, switch to oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects) or switch to or combine with oral gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects).

Third-line treatment 1.1.15 If satisfactory pain reduction is not achieved with second-line treatment:

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 12 of 150 refer the person to a specialist pain service and/or a condition- specific service7 and while waiting for referral: consider oral tramadol as third-line treatment instead of or in combination8 with the second-line treatment (For dosages please see box 1 Drug dosages). consider a topical lidocaine patch for treatment of localised pain for people who are unable to take oral medication because of medical conditions and/or disability.

Other treatments 1.1.16 Do not start treatment with a topical capsaicin 8% patch or with opioids (such as morphine or oxycodone) other than tramadol without an assessment by a specialist pain service or a condition- specific service7.

1.1.17 Pharmacological treatments other than those recommended in this guideline that are started by a specialist pain service or a condition- specific service7 may continue to be prescribed in non-specialist settings, with a multidisciplinary care plan, local shared care agreements and careful management of adverse effects.

Box 1 Drug dosages Start at a low dose, as indicated in the table. Titrate upwards to an effective dose or the person’s maximum tolerated dose (no higher than the maximum dose listed in the table) Higher doses should be considered in consultation with a specialist pain service.

Drug Starting dose Maximum dose Amitriptyline* 10 mg/day 75 mg/day

7 A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. 8 The combination of tramadol with amitriptyline, nortriptyline, imipramine or duloxetine is associated with a low risk of serotonin syndrome (the features of which include confusion, delirium, shivering, sweating, changes in blood pressure and myoclonus).

Neuropathic pain: NICE clinical guideline DRAFT September 2011 13 of 150 Gabapentin 300 mg once daily on day 1, then 3600 mg/day 300 mg twice daily on day 2, then 300 mg 3 times daily on day 3 or initially 300 mg 3 times daily on day 1c Pregabalin 150 mg/day a (divided into 2 600 mg/day (divided into 2 doses) doses) Duloxetine 60 mg/day a 120 mg/day Tramadol b 50-100mg not more often than 400 mg/day every 4 hours * Not licensed for this indication at time of publication (December 2011). Informed consent should be obtained and documented. a A lower starting dose may be appropriate for some people. b As monotherapy. More conservative titration may be required if used as combination therapy. c A less rapid escalation schedule may be more appropriate.

Neuropathic pain: NICE clinical guideline DRAFT September 2011 14 of 150 After the diagnosis of neuropathic pain and appropriate management of the underlying condition(s).

1 2 Care pathway See box A for information on drug dosing at all stages

2 People with painful diabetic neuropathy People with other neuropathic pain conditions including post-herpetic 3 neuralgia

4 First-line treatment Consider referring the person to Offer oral duloxetine as first-line First-line treatment a specialist pain service and/or a treatment. If duloxetine is Offer oral amitriptyline* or gabapentin condition-specific service1 at any contraindicated, offer oral amitriptyline*. 5 stage, including at initial presentation and at the regular clinical reviews, if: 6 they have severe pain or pain significantly limits their daily activities and Perform: 7 participation2 or early clinical review (see box B) their underlying health regular clinical reviews (see box C). condition has deteriorated. 8

9 Unsatisfactory pain Satisfactory pain Satisfactory pain reduction but person reduction * In these recommendations, reduction at maximum cannot tolerate adverse events Continue treatment – drug names are marked with an tolerated dose If amitriptyline* results in satisfactory consider gradually 10 asterisk if they do not have UK Move to second-line pain reduction as first-line treatment reducing dose over marketing authorisation for the treatment but the person cannot tolerate the time if improvement is indication in question at the time adverse effects, consider oral sustained 11 of publication (September 2011). imipramine* or nortriptyline* as an Informed consent should be alternative. obtained and documented. If gabapentin results in satisfactory 12 pain reduction as first-line treatment but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects, consider 13 oral pregabalin as an alternative. continued on next page… Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 15 of 150

After first-line treatment options have been tried, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person

14 See box A for information on drug dosing at all stages

15 People with other neuropathic pain conditions People with painful diabetic neuropathy including post-herpetic neuralgia 16 Second-line treatment Second-line treatment 17 If first-line treatment was with duloxetine, switch to oral amitriptyline* (or If first-line treatment was with amitriptyline* (or imipramine* or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the nortriptyline*), switch to or combine with oral gabapentin (or person cannot tolerate the adverse effects) or switch to or combine with oral pregabalin as an alternative if gabapentin is effective but the gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or 18 person has difficulty adhering to the dosage schedule or cannot tolerate the cannot tolerate the adverse effects). adverse effects). If first-line treatment was with gabapentin (or pregabalin) If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with oral amitriptyline* (or imipramine* or 19 switch to or combine with gabapentin (or pregabalin as an alternative if gabapentin nortriptyline* as an alternative if amitriptyline is effective but is effective but the person has difficulty adhering to the dosage schedule or cannot the person cannot tolerate the adverse effects). tolerate the adverse effects). 20

Perform: Satisfactory pain reduction 21 Unsatisfactory pain reduction at maximum tolerated dose early clinical review (see Continue treatment – consider Refer the person to a specialist pain service and/or a box B) gradually reducing dose over a condition-specific service and regular clinical reviews time if improvement is 22 while waiting for referral: (see box C). sustained consider oral tramadol as third-line treatment instead of or in combinationb with the second-line treatment 23 consider a topical lidocaine patch for treatment of localised pain for people who are unable to take oral * In these recommendations, drug names are marked with an asterisk if they do not have UK medication because of medical conditions and/or marketing authorisation for the indication in question at the time of publication (September 24 disability. 2011). Informed consent should be obtained and documented. a A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology 25 and oncology services. b The combination of tramadol with amitriptyline, nortriptyline, imipramine or duloxetine is associated with a low risk of serotonin syndrome (the features of which include confusion, 26 delirium, shivering, sweating, changes in blood pressure and myoclonus).

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 16 of 150

Other treatments Do not start treatment with a topical capsaicin 8% patch or with opioids (such as morphine or oxycodone) other than 27 tramadol without an assessment by a specialist pain service or a condition-specific service11. Pharmacological treatments other than those recommended in this guideline that are started by a specialist pain service or a condition-specific service may continue to be prescribed in non-specialist settings, with a multidisciplinary care plan, local 28 shared care agreements and careful management of adverse effects. A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. 29

30

Box A Drug dosages Box B Early clinical review 31 Start at a low dose, as indicated in the table. After starting or changing a treatment, perform an Titrate upwards to an effective dose or the person’s maximum tolerated dose (no higher early clinical review of dosage titration, tolerability than the maximum dose listed in the table). and adverse effects to assess suitability of chosen 32 Drug Starting dose Maximum dose treatment. Amitriptyline* 10 mg/day 75 mg/day 33 Gabapentin 300 mg once daily on day 1, then 300 mg 3600 mg/day twice daily on day 2, then 300 mg three times Box C Regular clinical reviews daily on day 3 or initially 300 mg three times a Perform regular clinical reviews to assess and 34 daily on day 1 monitor effectiveness of chosen treatment. Include Pregabalin 150 mg/dayb (divided into two doses) 600 mg/day (divided assessment of: into two doses) pain reduction 35 Duloxetine 60 mg/dayb 120 mg/day adverse effects daily activities and participation (such as ability to Tramadol 50–100mg not more often than every 4 hoursc 400 mg/day work and drive) 36 * Not licensed for this indication at time of publication (September 2011). Informed consent mood (in particular, possible depression and/or should be obtained and documented. anxiety) a A less rapid escalation schedule may be more appropriate for some people. quality of sleep b 37 A lower starting dose may be appropriate for some people. overall improvement as reported by the person. c As monotherapy. More conservative titration may be needed if used as combination therapy. 38

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 17 of 150

39 3 Evidence review and recommendations

40 For details of how this guideline was developed see appendix 10.1 – 10.11.

41 3.1 Methodology

42 Based on the guideline scope, neuropathic pain is treated as a ‘blanket 43 condition’ in this guideline regardless of its aetiologies, unless there is 44 sufficient, valid and robust clinical and health economics evidence that shows 45 the clinical efficacy and cost effectiveness of a particular treatment for a 46 specific neuropathic pain condition. Hence, the structure of this guideline, the 47 categorisation of neuropathic pain conditions with relevant pharmacological 48 treatments, and analyses were based on this prior rationale.

49 The scope and protocols of studies included in this guideline, as well as the 50 methods for analysis and synthesis, are briefly summarised below. This will 51 provide overall information and brief explanation for the characteristics of all 52 evidence statements (except for the ‘Key principles of care’ section). in the 53 guideline for the following sections.

54 Population and conditions 55 Adults (aged 18 years old or older with neuropathic pain conditions. The 56 different neuropathic pain conditions that were included in this guideline are 57 listed in Table 3. Because the scope of this guideline is to provide 58 recommendations for neuropathic pain as a chronic condition, adults with pain 59 arising directly from trauma and surgical procedures for less than 3 months 60 were excluded.

61 Settings 62 Although the scope of this guideline is to provide recommendations for non- 63 specialist settings, studies conducted in pain specialist clinics were also 64 included because extrapolating the evidence to non-specialist settings is 65 appropriate.

66 Treatments and comparators 67 Table 4 lists the 34 different pharmacological treatments were considered for 68 neuropathic pain in the four main drug classes (antidepressants, anti- Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 18 of 150

69 epileptics, opioid analgesics and topical treatments). The guideline sought to 70 investigate:

71 the clinical efficacy of the individual listed 34 pharmacological treatments 72 as monotherapy (placebo-controlled trials) 73 the clinical efficacy of individual pharmacological treatments against each 74 other (head-to-head monotherapy comparative trials) 75 the clinical efficacy of combination therapy against monotherapy or other 76 combination therapy (head-to-head combination therapy comparative 77 trials).

78 Only randomised controlled trials of the interventions above were included in 79 this guideline.

80 Critical outcomes 81 The critical outcomes for meta-analysis, based on the globally accepted 82 Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials 83 (IMMPACT) recommendations (Dworkin et al. 2005; Dworkin et al. 2008), 84 were: at least 30% pain reduction; at least 50% pain reduction; patient- 85 reported global improvement/impression of pain; and adverse effects. Specific 86 adverse effects for each drug class were selected and agreed by the GDG 87 through survey questionnaires (see appendix 10.3A), based on their expert 88 knowledge and experience (including that of patient and carer members).

89 Literature search 90 Systematic literature searches were carried out to identify all randomised 91 controlled trials on the 34 different pharmacological treatments (listed in 92 Table 4) for neuropathic pain conditions (listed in Table 3). For full search 93 strategies please see appendix XXX.(Please note full search strategies will be 94 available on publication of the guideline).

95 Analysis and synthesis 96 For this guideline, meta-analysis was adopted as the analytical method for 97 analysing the evidence and the GRADE methodology was adopted to 98 synthesize and presented the results. Overall, a fixed-effects model meta-

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 19 of 150

99 analysis was carried out on the critical outcomes by individual 100 pharmacological treatments across all neuropathic pain conditions. However, 101 as previously stated, if evidence was sufficient, valid and robust, the meta- 102 analysis would be carried out by individual pharmacological treatments of 103 specific neuropathic conditions (for example, painful diabetic neuropathy, 104 post-herpetic neuralgia). If there was significant heterogeneity from the meta- 105 analysis, a random-effects model was adopted for the meta-analysis with 106 potential sources for the heterogeneity noted in the full GRADE profiles. The 107 outcome would be downgraded by one level due to ‘inconsistency’ as 108 described in the GRADE methodology. All results from the meta-analyses 109 (relative risk or risk ratio [RR] and absolute risk [AR], with 95% confidence 110 intervals [CI]) are presented in the summary profiles in the guideline with 111 subsequent evidence statements, and in the full GRADE profiles in the 112 appendices (for full GRADE profiles, see appendix 9). No studies were 113 excluded on the basis of outcomes reported.

114 For the completeness of the evidence base, included studies that did not 115 report the critical outcomes recommended by the IMMPACT 116 recommendations (at least 30% pain reduction; at least 50% pain reduction; 117 patient-reported global improvement; adverse effects) (Dworkin et al. 2005; 118 Dworkin et al. 2008) were summarised in evidence tables (see 119 appendix 10.10). These pain outcomes (other than the critical outcomes) are 120 referred to as ‘other reported pain outcomes’ in this guideline. The other 121 reported pain outcomes in the included studies are also presented in the 122 summary profiles in the guideline, subsequent evidence statements, and in 123 the full GRADE profiles in the appendices, with the outcome downgraded by 124 one level due to ‘indirectness’ as described in the GRADE methodology. 125 These other reported pain outcomes included mean pain relief score, mean 126 pain intensity score, mean change in pain relief score from baseline, mean 127 change in pain intensity score from baseline and mean change in daily pain 128 score.

129 Only evidence on the critical outcomes recommended by the IMMPACT 130 recommendations (at least 30% pain reduction; at least 50% pain reduction;

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 20 of 150

131 patient-reported global improvement; adverse effects) was used to generate 132 recommendations. However, if evidence on the critical outcomes for particular 133 pharmacological treatments was scarce or limited, evidence from other 134 reported pain outcomes was used to assist and generate discussion among 135 the GDG to reach consensus, but not as the sole basis for making 136 recommendations. For included studies that did not report either critical 137 outcomes or other pain outcomes, study characteristics were summarised in 138 the evidence tables only for information (see the evidence tables in 139 appendix 10.10 for full information on each included study).

140 For more details on the review protocols and specific inclusion and exclusion 141 criteria, please see appendix 10.2.

142 Table 3 Neuropathic pain conditions (search terms) included in the 143 searches Central neuropathic pain/central pain Compression neuropathies/nerve compression syndromes Facial neuralgia HIV-related neuropathy Idiopathic neuropathies Mixed neuropathic pain Multiple sclerosis Neurogenic pain Neuropathic cancer pain/cancer pain Neuropathic pain Painful diabetic neuropathy/diabetic neuropathy Peripheral nerve injury Peripheral neuropathies Phantom limb pain Post-amputation pain Post-herpetic neuralgia Post-stroke pain Post-treatment/post-surgery/post-operative pain Radiculopathies/radicular pain Spinal cord injury Trigeminal neuralgia 144

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 21 of 150

145 Table 4 Pharmacological treatments considered for the clinical guideline 146 on neuropathic pain Drug class: subclass Drug Antidepressants: tricyclic antidepressants (TCAs) Amitriptyline Clomipramine Desipramine Dosulepin (dothiepin) Doxepin Imipramine Lofepramine Nortriptyline Trimipramine Antidepressants: selective serotonin reuptake Citalopram inhibitors (SSRIs) Fluoxetine Paroxetine Sertraline Antidepressants: serotonin–norepinephrine reuptake Duloxetine inhibitors (SNRIs) Venlafaxine Anti-epileptics (anticonvulsants) Carbamazepine Gabapentin Lamotrigine Oxcarbazepine Phenytoin Pregabalin Sodium valproate Topiramate Opioid analgesics Buprenorphine Co-codamol Codeine phosphate Co-dydramol Dihydrocodeine Fentanyl Morphine Oxycodone Tramadol Topical treatments Topical capsaicin Topical lidocaine 147

148 3.1.1 Evidence Review

149 Review questions 150 Based on the scope and methodology set out in section 3.1 above, three 151 review questions were formulated. As stated in section 3.1 above, if sufficient, Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 22 of 150

152 valid and robust evidence showed the clinical efficacy of a particular treatment 153 for a specific neuropathic pain condition, evidence would be analysed based 154 on that particular neuropathic pain condition (for example, painful diabetic 155 neuropathy, post-herpetic neuralgia, and other neuropathic pain conditions) in 156 each review question below.

157 Review question 1 158 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 159 analgesics, topical lidocaine and topical capsaicin as monotherapy (against 160 placebo) for the management of neuropathic pain conditions in adults in non- 161 specialist settings?

162 Review question 2 163 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 164 analgesics, topical lidocaine and topical capsaicin as combination therapy 165 (against monotherapy or other combination therapy) for the management of 166 neuropathic pain in adults in non-specialist settings?

167 Review question 3 168 What is the clinical efficacy of antidepressants, compared with anti-epileptics, 169 opioid analgesics, topical lidocaine and topical capsaicin (and vice-versa) as 170 monotherapy for the management of neuropathic pain in adults in non- 171 specialist settings?

172 Overall summary of evidence 173 A total of 29,237 studies were retrieved by the systematic searches for the 174 guideline (antidepressants = 3641, anti-epileptics = 6167, opioid analgesics = 175 12,075, topical capsaicin and topical lidocaine = 7196, neurogenic pain = 176 158). From the 29,237 studies, 101 randomised placebo-controlled trials of 177 monotherapy, 14 head-to-head comparative trials of monotherapy and 178 combination therapy were included, based on the inclusion and exclusion 179 criteria suggested by the GDG through two short survey questionnaires9. The

9 For the full search strategies, see appendix 10.7; for the two GDG short questionnaires on inclusion and exclusion criteria, see appendix 10.3A; for the full review protocol, see appendix 10.2. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 23 of 150

180 searches did not identify any placebo-controlled studies that met the inclusion 181 and exclusion criteria for 15 of the pharmacological treatments (see table 5). 182 The 115 included studies are summarised in table 6.

183 Table 5 Pharmacological treatments for which no studies met the 184 inclusion and exclusion criteria Drug class: subclass Drug Antidepressants: tricyclic antidepressants Dosulepin (dothiepin) (TCAs) Doxepin Lofepramine Trimipramine Antidepressants: selective serotonin reuptake Citalopram inhibitors (SSRIs) Fluoxetine Paroxetine Sertraline Anti-epileptics (anticonvulsants) Phenytoin Opioid analgesics Buprenorphine Co-codamol Codeine phosphate Co-dydramol Dihydrocodeine Fentanyl 185

186 Table 6 Summary of included randomised placebo-controlled trials on 187 antidepressants, anti-epileptics, opioid analgesics and topical 188 treatments, and head-to-head comparative and combination therapy 189 trials, for the treatment of neuropathic pain Drug class No. of Treatment Key outcomes studies included Antidepressants 11 Amitriptyline 30%, Global, mean pain intensity (TCAs) score, mean pain relief scores, AEs Antidepressants 2 Desipramine Global, AEs (TCAs) Antidepressants 1 Nortriptyline Global (TCAs) Antidepressants 1 Imipramine Global, AEs (TCAs) Antidepressants 5 Duloxetine 30%, 50%, AEs (SNRIs) Antidepressants 4 Venlafaxine 50%, Global, mean pain intensity (SNRIs) score, AEs

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 24 of 150

Subtotal 24 Anti-epileptics 2 Carbamazepine Global Anti-epileptics 3 Oxcarbazepine 30%, 50%, Global, mean pain relief score, AEs Anti-epileptics 3 Sodium Mean pain relief score, mean pain valproate intensity score, AEs Anti-epileptics 3 Topiramate 30%, 50%, Global, AEs Anti-epileptics 10 Lamotrigine 30%, 50%, Global, AEs Anti-epileptics 13 Gabapentin 30%, 50%, Global, mean change in pain intensity score, mean pain relief score, AEs Anti-epileptics 16 Pregabalin 30%, 50%, Global, mean pain intensity score, AEs Subtotal 50 Opioid analgesics 5 Tramadol 50%, mean pain intensity score, AEs Opioid analgesics 3 Morphine 30%, 50%, Global, AEs Opioid analgesics 1 Oxycodone Mean change in pain intensity score, AEs Subtotal 9 Topical treatments 13 Topical 40%, 50%, Global, mean pain relief capsaicin score, mean change in pain intensity score, mean change in pain relief score, AEs Topical treatments 5 Topical Mean pain relief score, mean pain lidocaine intensity score, mean change in pain relief score, mean change in pain intensity score, AEs Subtotal 18 Antidepressants vs 3 Amitriptyline vs 30%, Global, AEs, mean change in anti-epileptics gabapentin pain intensity score, mean change in pain relief score Antidepressants vs 1 Amitriptyline vs 50%, Global, AEs anti-epileptics pregabalin Antidepressants vs 1 Nortriptyline vs 50%, mean change in pain relief anti-epileptics gabapentin score, AEs Antidepressants vs 1 Amitriptyline vs Global, AEs anti-epileptics carbamazepine Antidepressants vs 1 Amitriptyline vs Mean change in pain relief score, topical capsaicin topical capsaicin mean change in pain intensity score, AEs Anti-epileptics vs 1 Pregabalin vs 30%, 50%, Global, AEs topical lidocaine topical lidocaine Antidepressants vs 1 Amitriptyline vs AEs antidepressants nortriptyline Antidepressants vs 1 Imipramine vs Global, AEs antidepressants venlafaxine Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 25 of 150

Anti-epileptics + 1 Gabapentin + Mean pain relief score, AEs opioids vs anti- oxycodone vs epileptics gabapentin Anti-epileptics + 1 Pregabalin + Mean pain intensity score, AEs opioids vs anti- oxycodone vs epileptics vs opioids pregabalin vs oxycodone Anti-epileptics + 1 Gabapentin + Mean change in daily pain score antidepressants vs nortriptyline vs anti-epileptics vs gabapentin vs antidepressants nortriptyline Antidepressants + anti- 1 Amitriptyline + 50% epileptics vs pregabalin vs antidepressants vs amitriptyline vs anti-epileptics pregabalin Subtotal 14 TOTAL 115 TCA = tricyclic antidepressant; SNRI = serotonin–norepinephrine reuptake inhibitor; 30% = at least 30% pain reduction; 40% = at least 40% pain reduction; 50% = at least 50% pain reduction; Global = patient-reported global improvement; AEs = adverse effects. 190

191 Structure of the guideline and analyses 192 From the 115 included studies, nearly 50% were on two specific neuropathic 193 pain conditions, namely painful diabetic neuropathy (PDN; 34 studies) and 194 post-herpetic neuralgia (PHN; 21 studies). Consequently, these two specific 195 neuropathic conditions were perceived as having sufficient and robust 196 evidence and hence meta-analyses were carried out for individual 197 pharmacological treatments for PDN and PHN.

198 The other 60 included studies are on various other neuropathic pain 199 conditions (for example, spinal cord injury, neuropathic cancer pain, HIV- 200 related neuropathy, post-stroke neuropathic pain, phantom limb pain, central 201 pain, radiculopathy, polyneuropathy, post-traumatic neuropathic pain including 202 post-surgical neuropathic pain, or mixed neuropathic pain). Because none of 203 these neuropathic pain conditions had significant numbers of studies available 204 (unlike like PDN and PHN), a meta-analysis was done of the 60 included 205 studies as ‘other neuropathic pain conditions’ for each drug.

206 Issues on key principles of care were also discussed after decisions were 207 made on which pharmacological treatments should be recommended. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 26 of 150

208 Therefore, the guideline is structured under three sections as follow: PDN, 209 PHN, and other neuropathic pain conditions, with discussion on key principles 210 of care for all neuropathic pain conditions combined.

211 The structure of the guideline is based on the chronological order of how the 212 GDG assessed and discussed the evidence to allow readers to understand 213 the rationales and decisions made at each stage. However, the structure of 214 the guideline does not match the order of the recommendations because the 215 GDG felt that recommendations need to be ordered to aid implementation, 216 rather than by the order the evidence was analysed.

217 3.1.2 Health economics

218 A search was conducted for published health economic analysis and no 219 appropriate publications were identified. Full details are presented in appendix 220 10.11 including reasons for exclusion. However, the GDG had access to a 221 relevant in development health technology assessment (HTA) report that is 222 due for publication after guideline development had ended. This draft HTA 223 report by Fox-Rushby et al (Project abstract available from 224 www.hta.ac.uk/1527) reviewed the clinical and cost effectiveness of different 225 treatment pathways for neuropathic pain. A overview is presented in appendix 226 10.11. Below is a discussion on the applicability of this study to the clinical 227 guideline.

228 Applicability of the HTA model to the guideline 229 It is recognised that the methodology adopted for the draft HTA report, in 230 relation to both the efficacy review and the health economic evaluation, was 231 systematic and of high quality. Therefore, this discussion will contrast the 232 approaches used for the draft HTA report and the current clinical guideline 233 and consider their potential impact on interpretation and generalisability for 234 this guideline. Then the remaining limitations of the model will be discussed.

235 Decision problem 236 The draft HTA report reviewed the evidence on only two conditions, namely 237 PHN and PDN. Other sub-populations were considered difficult to model

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 27 of 150

238 because of lack of data, as shown by the effectiveness and economic reviews 239 of the literature conducted for this guideline.

240 The drugs covered by the draft HTA report differed to those included in this 241 guideline. In particular, amitriptyline was excluded from the economic analysis 242 because the available evidence was poor. In addition, the analysis included a 243 number of treatments that the GDG considered were unsuitable for primary 244 care such as venlafaxine and epidural methylprednisolone. The doses used in 245 the model may not be representative of clinical practice and therefore may 246 reduce the generalisability of the results. In addition, the model could not 247 consider combination treatment and therefore, could not be used to inform a 248 clinical pathway. However, the GDG were able to combine the results of the 249 economic analysis with their clinical opinion within the framework of the 250 guideline to produce recommendations.

251 Clinical data 252 The clinical data included in the HTA model were based on a systematic 253 search for randomised controlled trial (RCT) data completed in 2009. The 254 search strategy and included studies differed from those in the clinical 255 evidence review in this guideline, which included a number of new head-to- 256 head studies and data for the recently licensed capsaicin 8% patch.

257 The reliance of the HTA model on data from clinical trials means that it is 258 susceptible to the weaknesses associated with trials, such as failing to reflect 259 real clinical practice. In particular, the drug doses were modelled as 260 prescribed in the clinical trials. However, drug doses in trials do not 261 necessarily reflect doses prescribed in practice, which may be substantially 262 higher. This is an important issue and affects the evidence of both clinical and 263 cost effectiveness.

264 In addition, the data on minor adverse events are possibly unrepresentative. 265 In a drug trial, a patient experiencing minor adverse events may be asked to 266 continue to take the drug for the short trial duration. By contrast, a member of 267 the public under the care of their GP and/or a specialist may agree to try an

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 28 of 150

268 alternative drug in the hope of obtaining pain relief without unpleasant adverse 269 events.

270 Comorbidities associated with PDN and diabetes, such as cardiovascular 271 disease and peripheral vascular disease, were not accounted for in the model 272 because the systematic review excluded efficacy trials that included patients 273 with comorbidities.

274 Quality of life 275 For the health economic modelling in the draft HTA report, pain relief was 276 used to define the health states, to which a global valuation of quality of life 277 was assigned – that is, a utility estimate. Pain and other outcome data are 278 commonly used to feed into utility estimates, and pain is a dimension on the 279 EQ-5D tool that is frequently used to measure quality of life for economic 280 evaluations.

281 If a drug does not provide a 50% pain reduction or more, then it does not incur 282 any health benefits in the HTA model. However, introducing a lower cut-off 283 point such as 30% pain reduction, could result in some benefit, albeit smaller 284 than that obtained with a drug that reduces pain by at least 50%. Thus the 285 differences between the more effective and less effective drugs may become 286 smaller with this lower cut-off added, but this is unlikely to change the ordering 287 of the treatments in the analysis.

288 Resource use and costs 289 For both the PHN and PDN models, expert opinion supplemented the data if 290 insufficient published data were available to populate the model. Six experts in 291 PHN and four experts in PDN completed a questionnaire, and the answers 292 obtained informed the costing, as well as providing information on adverse 293 events.

294 The care pathways used in the deterministic and probabilistic modelling do not 295 appear to match the definition of ‘non-specialist settings’ used in the current 296 guideline. This has two possible implications: first, cost estimates may not 297 reflect those relevant for the current guideline; second, the drugs may not be 298 suitable to be prescribed in a non-specialist setting. For example, healthcare Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 29 of 150

299 professionals who are not pain specialists may have different levels of 300 experience and confidence in prescribing and managing the long-term use of 301 opioids.

302 Conclusion 303 The draft HTA report broadly overlaps with the guideline’s clinical questions; 304 however, there are a number of significant differences from the guideline. This 305 limits the generalisability of the HTA model’s results. The GDG concluded that 306 the HTA analysis was partially applicable and had minor limitations. 307 Therefore, the outputs of the HTA report will be considered alongside the 308 clinical evidence, information on acquisition costs and GDG experience when 309 assessing the cost effectiveness of treatments.

310 3.2 Painful diabetic neuropathy (PDN)

311 3.2.1 Review questions

312 Review question 1 313 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 314 analgesics, topical lidocaine and topical capsaicin as monotherapy (against 315 placebo) for the management of neuropathic pain condition (painful diabetic 316 neuropathy) in adults in non-specialist settings?

317 Review question 2 318 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 319 analgesics, topical lidocaine and topical capsaicin as combination therapy 320 (against monotherapy or other combination therapy) for the management of 321 neuropathic pain (painful diabetic neuropathy) in adults in non-specialist 322 settings?

323 Review question 3 324 What is the clinical efficacy of antidepressants, compared with anti-epileptics, 325 opioid analgesics, topical lidocaine and topical capsaicin (and vice-versa) as 326 monotherapy for the management of neuropathic pain (painful diabetic 327 neuropathy) in adults in non-specialist settings?

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 30 of 150

328 3.2.2 Evidence review

329 A total of 34 randomised controlled trials were included for painful diabetic 330 neuropathy (PDN). Of the 34 listed included pharmacological treatments in 331 (Table 4), no study was identified or met the inclusion and exclusion criteria 332 for the following pharmacological treatments (see table 7).

333 For the characteristics of included studies please see Tables 8–12.

334 Table 7 Pharmacological treatments for which no study was identified or 335 met the inclusion and exclusion criteria for PDN Drug class: subclass Drug Antidepressants: tricyclic antidepressants Clomipramine (TCAs) Dosulepin (dothiepin) Doxepin Imipramine Lofepramine Nortriptyline Trimipramine Antidepressants: selective serotonin reuptake Citalopram inhibitors (SSRIs) Fluoxetine Paroxetine Sertraline Anti-epileptics (anticonvulsants) Carbamazepine Phenytoin Opioid analgesics Buprenorphine Co-codamol Codeine phosphate Co-dydramol Dihydrocodeine Fentanyl Morphine Topical treatments Topical lidocaine 336

337 Table 8 Characteristics of included studies for PDN: antidepressants as 338 monotherapy (placebo-controlled trials)

Author Study Condition Treatment Titration or Mean Outcomes period (oral) fixed dosage dose (mg/day) (mg/day) Max et al. (1991) 6 weeks PDN Desipramine 12.5–250 201 Global, AEs Goldstein et al. 12 weeks PDN Duloxetine 20, 60, 120 N/A 50%, AEs (2005) Raskin et al. 12 weeks PDN Duloxetine 60, 120 N/A 50%, AEs (2005) Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 31 of 150

Wernicke et al. 12 weeks PDN Duloxetine 60, 120 N/A 30%, 50%, AEs (2006) Gao et al. (2010) 12 weeks PDN Duloxetine 30-120 N/A 30%, 50%, AEs Rowbotham et al. 6 weeks PDN Venlafaxine 75, 150–225 N/A 50%, AEs (2004) PDN = painful diabetic neuropathy; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects; N/A = not applicable. 339

340 Table 9 Characteristics of included studies for PDN: anti-epileptics as 341 monotherapy (placebo-controlled trials)

Author Study Condition Treatment (oral) Titration or Outcomes period fixed dosage (mg/day) Beydoun et al. 16 weeks PDN Oxcarbazepine to 600 Global, AEs (2006) Dogra et al. 16 weeks PDN Oxcarbazepine 300–1800 30%, 50%, Global, (2005) AEs Grosskopf et al. 16 weeks PDN Oxcarbazepine 300–600 Mean pain relief (2006) score, AEs 3 months PDN Sodium valproate 20 per kg Mean pain Agrawal et al. intensity score, (2009) AEs Kochar et 4 weeks PDN Sodium valproate 1200 AEs al.(2002) Kochar et al. 3 months PDN Sodium valproate 500 Mean pain relief (2004) score, AEs Raskin et al. 12 weeks PDN Topiramate 25–400 30%, 50%, Global, (2004) AEs Thienel et al. 22 weeks PDN Topiramate 100, 200, 400 AEs (2004) Eisenberg et al. 8 weeks PDN Lamotrigine 25–400 50%, Global, AEs (2001) Luria et al. 8 weeks PDN Lamotrigine 25–400 50%, AEs (2000) Vinik et al. 19 weeks PDN Lamotrigine 200, 300, 400 30%, 50%, AEs (2007) Backonja et al. 8 weeks PDN Gabapentin to 3600 Global, AEs (1998) Simpson (2001) 8 weeks PDN Gabapentin to 3600 Global, AEs 13 weeks PDN Pregabalin to 600 Mean pain Arezzo et intensity score, al.(2008) AEs Lesser et al. 5 weeks PDN Pregabalin to 75, 300, 600 30%, 50%, Global, (2004) AEs Richter et al. 6 weeks PDN Pregabalin 25–150, 100–600 50%, AEs (2005) Rosenstock et al. 8 weeks PDN Pregabalin 300 50%, AEs (2004) Tölle et al. 12 weeks PDN Pregabalin 150, 300, 50%, Global, AEs (2008) 300/600 Satoh et al. 14 weeks PDN Pregabalin 300, 600 50%, AEs (2011) PDN = painful diabetic neuropathy; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 342

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 32 of 150

343 Table 10 Characteristics of included studies for PDN: opioid analgesics 344 as monotherapy (placebo-controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (oral) dosage (mg/day) Harati et al. 4 weeks PDN Tramadol 200–400 Mean pain intensity (1998) score, AEs Gimbel et al. 6 weeks PDN Oxycodone 10–120 Mean change in pain (2003) intensity score, AEs PDN = painful diabetic neuropathy; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 345

346 Table 11 Characteristics of included studies for PDN: topical capsaicin 347 and topical lidocaine as monotherapy (placebo-controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (topical) dosage (times/day) Scheffler et 8 weeks PDN Capsaicin 0.075% cream, 4 Mean pain relief al. (1991) score, mean change in pain intensity score, AEs Tandan et al. 8 weeks PDN Capsaicin 0.075% cream, 4 Global, AEs (1992) PDN = painful diabetic neuropathy; Global = patient-reported global improvement; AEs = adverse effects. 348

349 Table 12 Characteristics of included studies for PDN: comparative trials 350 and combination therapy (randomised controlled trials) Author Study Condition T1 T2 Titration or Outcomes period fixed dosage (mg/day) Cross-class head-to-head comparison TCAs vs anti-epileptics Morello et 6 weeks PDN Amitriptyline Gabapentin Ami: 25–75 Global, mean al. (1999) Gaba: 900–1800 change in pain intensity score, AEs Dallocchio 12 PDN Amitriptyline Gabapentin Ami: 10–90 Mean change et al. weeks Gaba: 400–2400 in pain relief (2000) score, AEs Bansal et 5 weeks PDN Amitriptyline Pregabalin Ami: 10-50 Global, 50%, al. (2009) Pre: 150-600 AEs TCAs vs topical capsaicin Biesbroeck 8 weeks PDN Amitriptyline Topical Ami: 25–125 Mean change et al. capsaicin Cap: 0.075% in pain relief (1995) cream, 4 score, mean times/day change in pain intensity score, AEs Combination therapy Anti-epileptics + opioids vs anti-epileptics Hanna et 12 PDN Gabapentin Gabapentin Gaba: 600–1800 Mean pain al.(2008) weeks + oxycodone Oxy: 5–80 relief score, AEs

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 33 of 150

Author Study Condition T1 T2 Titration or Outcomes period fixed dosage (mg/day) T1 = treatment 1; T2 = treatment 2; PDN = painful diabetic neuropathy; Global = patient-reported global improvement; 50% = at least 50% pain reduction; AEs = adverse effects. 351

352 Summary profiles 353 Meta-analyses were conducted based on the methodology stated in 354 section 3.1 and presented in the following summary profiles based on 355 individual pharmacological treatments (for full GRADE profiles, see appendix 356 10.9).

357 Antidepressants 358 Table 13 Summary profile – desipramine as monotherapy (placebo- 359 controlled trials) No of Relative risk Desipramine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 11 11/20 2/20 RR 5.50 (1.39 to 45 more per 100 (from 4 (55%) (10%) 21.71) more to 100 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 6 weeks) 11 2/20 0/20 RR 5.00 (0.26 to – (10%) (0%) 98.00) VERY LOW Primary outcome: dry mouth (adverse effects) (follow-up 6 weeks) 11 8/20 9/20 RR 0.89 (0.43 to 1.83) 5 fewer per 100 (from 26 (40%) (45%) fewer to 37 more) VERY LOW Primary outcome: sedation (adverse effects) (follow-up 6 weeks) 11 8/20 8/20 RR 1.00 (0.47 to 2.14) 0 fewer per 100 (from 21 (40%) (40%) fewer to 46 more) VERY LOW Primary outcome: any adverse effects: unspecified (follow-up 6 weeks) 11 18/20 17/20 RR 1.06 (0.84 to 1.34) 5 more per 100 (from 14 (90%) (85%) fewer to 29 more) VERY LOW 1 Max et al. (1991). 360

361 Table 14 Summary profile – duloxetine as monotherapy (placebo- 362 controlled trials) No of Relative risk Duloxetine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 12 weeks) 21 220/327 111/215 RR 1.33 (0.95 to 1.88) 17 more per 100 (from 3 (67.3%) (51.6%) fewer to 45 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 12 weeks) 42 485/896 164/443 RR 1.51 (1.17 to 1.94) 19 more per 100 (from 6 (54.1%) (37%) more to 35 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 12 weeks) 42 113/906 21/448 RR 2.63 (1.68 to 4.12) 8 more per 100 (from 3 Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 34 of 150

(12.5%) (4.7%) more to 15 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 12 weeks) 33 90/674 26/332 RR 1.81 (1.17 to 2.79) 6 more per 100 (from 1 (13.4%) (7.8%) more to 14 more) MODERATE Primary outcome: dry mouth (adverse effects) (follow-up 12 weeks) 24 37/448 10/224 RR 1.61 (0.82 to 3.20) 3 more per 100 (from 1 (8.3%) (4.5%) fewer to 10 more) LOW Primary outcome: gastrointestinal disturbances (adverse effects) (follow-up 12 weeks) 25 28/332 8/217 RR 2.53 (1.13 to 5.67) 6 more per 100 (from 0 (8.4%) (3.7%) more to 17 more) LOW Primary outcome: vomiting (adverse effects) (follow-up 12 weeks) 16 6/106 5/109 RR 1.23 (0.39 to 3.92) 1 more per 100 (from 3 (5.7%) (4.6%) fewer to 13 more) VERY LOW Primary outcome: any adverse effects: unspecified (follow-up 12 weeks) 16 86/106 78/109 RR 1.13 (0.98 to 1.32) 9 more per 100 (from 1 (81.1%) (71.6%) fewer to 23 more) MODERATE 1 Gao et al. (2010); Wernicke et al. (2006). 2 Gao et al. (2010); Goldstein et al. (2005); Raskin et al. (2005); Wernicke et al. (2006). 3 Gao et al. (2010); Goldstein et al. (2005); Wernicke et al. (2006). 4 Gao et al. (2010); Goldstein et al. (2005). 5 Gao et al. (2010); Wernicke et al. (2006). 6 Gao et al. (2010). 363

364 Table 15 Summary profile – venlafaxine as monotherapy (placebo- 365 controlled trials) No of Relative risk Venlafaxine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 50% pain reduction (follow-up 6 weeks) 11 77/163 27/80 RR 1.40 (0.99 to 1.98) 13 more per 100 (from 0 (47.2%) (33.8%) fewer to 33 more) MODERATE Primary outcome: vomiting (adverse effects) (follow-up 6 weeks) 11 9/164 0/81 RR 9.44 (0.56 to – (5.5%) (0%) 160.24) VERY LOW 1 Rowbotham et al. (2004). 366

367 Anti-epileptics 368 Table 16 Summary profile – gabapentin as monotherapy (placebo- 369 controlled trials) No of Relative risk Gabapentin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 8 weeks) 21 62/106 32/103 RR 1.88 (1.35 to 2.61) 27 more per 100 (from 11 (58.5%) (31.1%) more to 50 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 8 weeks) 21 8/114 6/111 RR 1.29 (0.46 to 3.60) 2 more per 100 (from 3 (7%) (5.4%) fewer to 14 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 8 weeks) 21 26/111 5/108 RR 5.05 (2.02 to 19 more per 100 (from 5 (23.4%) (4.6%) 12.67) more to 54 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 8 weeks) 21 25/111 6/108 RR 4.05 (1.73 to 9.47) 17 more per 100 (from 4 (22.5%) (5.6%) more to 47 more) LOW Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 35 of 150

1 Backonja et al (1998); Simpson et al. (2001). 370

371 Table 17 Summary profile – pregabalin as monotherapy (placebo- 372 controlled trials) No of Relative risk Pregabalin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 5 weeks) 11 103/162 32/97 RR 1.93 (1.42 to 2.62) 31 more per 100 (from 14 (63.6%) (33%) more to 53 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 5 to 14 weeks) 52 298/808 91/459 RR 1.87 (1.33 to 2.63) 17 more per 100 (from 7 (36.9%) (19.8%) more to 32 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 5 to 12 weeks) 23 208/396 49/174 RR 1.89 (1.04 to 3.45) 25 more per 100 (from 1 (52.5%) (28.2%) more to 69 more) LOW Primary outcome: number of withdrawals due to adverse effects (follow-up 5 to 14 weeks) 64 97/1039 28/569 RR 2.13 (1.40 to 3.23) 6 more per 100 (from 2 (9.3%) (4.9%) more to 11 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 5 to 14 weeks) 64 222/1039 31/569 RR 4.53 (3.14 to 6.54) 19 more per 100 (from 12 (21.4%) (5.4%) more to 30 more) MODERATE Primary outcome: somnolence (adverse effects) (follow-up 5 to 14 weeks) 64 155/1039 26/569 RR 3.71 (2.46 to 5.58) 12 more per 100 (from 7 (14.9%) (4.6%) more to 21 more) MODERATE Primary outcome: weight gain (adverse effects) (follow-up 6 to 14 weeks) 45 60/723 4/402 RR 7.82 (3.12 to 7 more per 100 (from 2 (8.3%) (1%) 19.60) more to 19 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 to 14 weeks) 26 159/257 68/206 RR 1.87 (1.50 to 2.32) 29 more per 100 (from 17 (61.9%) (33%) more to 44 more) MODERATE 1 Lesser et al. (2004). 2 Lesser et al. (2004); Richter et al. (2005); Rosenstock et al. (2004); Tölle et al. (2008); Satoh et al. (2011). 3 Lesser et al. (2004); Tölle et al. (2008). 4 Arezzo et al. (2008); Lesser et al. (2004); Richter et al. (2005); Rosenstock et al. (2004); Tölle et al. (2008); Satoh et al. (2011). 5 Arezzo et al. (2008); Richter et al. (2005); Tölle et al. (2008); Satoh et al. (2011). 6 Rosenstock et al. (2004); Sath et al. (2011). 373

374 Table 18 Summary profile – lamotrigine as monotherapy (placebo- 375 controlled trials) No of Relative risk Lamotrigine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 19 weeks) 21 110/324 41/120 RR 1.00 (0.74 to 1.33) 0 fewer per 100 (from 9 (34%) (34.2%) fewer to 11 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 8 to 19 weeks) 32 92/351 35/146 RR 1.13 (0.81 to 1.57) 3 more per 100 (from 5 (26.2%) (24%) fewer to 14 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 8 weeks) 13 7/22 2/21 RR 3.34 (0.78 to 22 more per 100 (from 2 (31.8%) (9.5%) 14.29) fewer to 100 more) MODERATE Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 36 of 150

Primary outcome: number of withdrawals due to adverse effects (follow-up 8 to 19 weeks) 44 72/579 16/220 RR 1.58 (0.94 to 2.66) 4 more per 100 (from 0 (12.4%) (7.3%) fewer to 12 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 8 to 19 weeks) 32 43/559 12/200 RR 1.40 (0.73 to 2.68) 2 more per 100 (from 2 (7.7%) (6%) fewer to 10 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 to 19 weeks) 32 415/559 137/200 RR 1.05 (0.89 to 1.25) 3 more per 100 (from 8 (74.2%) (68.5%) fewer to 17 more) HIGH 1 Vinik et al. (2007a); Vinik et al. (2007b). 2 Eisenberg et al. (2001); Vinik et al. (2007a); Vinik et al. (2007b). 3 Eisenberg et al. (2001) .4 Eisenberg et al. (2001); Luria et al. (2000); Vinik et al. (2007a); Vinik et al. (2007b). 376

377 Table 19 Summary profile – topiramate as monotherapy (placebo- 378 controlled trials) No of Relative risk Topiramate Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 12 weeks) 11 103/208 37/109 RR 1.46 (1.09 to 1.96) 16 more per 100 (from 3 (49.5%) (33.9%) more to 33 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 12 weeks) 11 74/208 23/109 RR 1.69 (1.12 to 2.53) 15 more per 100 (from 3 (35.6%) (21.1%) more to 32 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 12 weeks) 11 112/208 37/109 RR 1.59 (1.19 to 2.12) 20 more per 100 (from 6 (53.8%) (33.9%) more to 38 more) MODERATE Primary outcome: no. of withdrawals due to adverse effects (follow-up 12 to 22 weeks) 22 265/1099 41/493 RR 2.90 (2.12 to 3.96) 16 more per 100 (from 9 (24.1%) (8.3%) more to 25 more) HIGH Primary outcome: dizziness (adverse effects) (follow-up 12 weeks) 11 15/211 6/109 RR 1.29 (0.52 to 3.23) 2 more per 100 (from 3 (7.1%) (5.5%) fewer to 12 more) VERY LOW Primary outcome: somnolence (adverse effects) (follow-up 12 to 22 weeks) 22 108/1096 19/493 RR 2.56 (1.59 to 4.11) 6 more per 100 (from 2 (9.9%) (3.9%) more to 12 more) MODERATE Primary outcome: fatigue (adverse effects) (follow-up 12 to 22 weeks) 22 158/1096 44/493 RR 1.58 (1.15 to 2.16) 5 more per 100 (from 1 (14.4%) (8.9%) more to 10 more) MODERATE Primary outcome: any adverse effects: unspecified (follow-up 12 weeks) 11 170/214 77/109 RR 1.12 (0.98 to 1.29) 8 more per 100 (from 1 (79.4%) (70.6%) fewer to 20 more) MODERATE 1 Raskin et al. (2004). 2Raskin et al. (2004); 3 Thienel et al. (2004). 379

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 37 of 150

380 Table 20 Summary profile – oxcarbazepine as monotherapy (placebo- 381 controlled trials) No of Relative risk Oxcarbazepine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 16 weeks) 11 31/69 22/77 RR 1.57 (1.01 to 2.44) 16 more per 100 (from 0 (44.9%) (28.6%) more to 41 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 16 weeks) 11 24/69 14/77 RR 1.91 (1.08 to 3.39) 17 more per 100 (from 1 (34.8%) (18.2%) more to 43 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 16 weeks) 22 97/229 52/149 RR 1.16 (0.90 to 1.49) 6 more per 100 (from 3 (42.4%) (34.9%) fewer to 17 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 16 weeks) 33 102/398 16/236 RR 3.83 (2.29 to 6.40) 19 more per 100 (from 9 (25.6%) (6.8%) more to 37 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 16 weeks) 22 58/310 3/159 RR 8.90 (2.81 to 15 more per 100 (from 3 (18.7%) (1.9%) 28.24) more to 51 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 16 weeks) 22 21/310 3/159 RR 2.95 (1.04 to 8.35) 4 more per 100 (from 0 (6.8%) (1.9%) more to 14 more) LOW Primary outcome: fatigue (adverse effects) (follow-up 16 weeks) 22 31/310 7/159 RR 1.83 (0.83 to 4.00) 4 more per 100 (from 1 (10%) (4.4%) fewer to 13 more) LOW 1 Dogra et al. (2005). 2 Dogra et al. (2005); Beydoun et al. (2006). 3 Dogra et al. (2005); Beydoun et al. (2006); Grosskopf et al. (2006). 382

383 Table 21 Summary profile – sodium valproate as monotherapy (placebo- 384 controlled trials) No of Sodium Relative risk Placebo Absolute risk Quality studies valproate (95% CI) Primary outcome: number of withdrawals owing to adverse effects (follow-up 4 to 12 weeks) 21 2/52 0/51 RR 2.93 (0.32 to – (3.8%) (0%) 27.29) LOW Primary outcome: any adverse effects: unspecified (follow-up 12 weeks) 12 4/20 1/20 RR 4.00 (0.49 to 15 more per 100 (from 3 (20%) (5%) 32.72) fewer to 100 more) VERY LOW Other reported pain outcome: pain intensity (scale: VASpi-10 cm) (follow-up 12 weeks) 12 20 20 Treatment = 6.2 (1.4); Placebo = 6.9 (1.0) p > 0.05 LOW Other reported pain outcome: pain relief (scale: VASpr-100 mm) (follow-up 12 weeks) 13 22 21 Treatment = 30.0 (99.4); Placebo = 60.0 (84.2) p < 0.001 LOW 1 Kochar et al. (2002); Kochar et al. (2004). 2 Agrawal et al. (2009). 3 Kochar et al. (2004). 385

386 Note: no study on sodium valproate that reported the critical outcomes on pain was identified or met the 387 inclusion and exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 38 of 150

388 Opioid analgesics 389 Table 22 Summary profile – tramadol as monotherapy (placebo- 390 controlled trials) No of Relative risk Tramadol Placebo Absolute risk Quality studies (95% CI) Primary outcome: withdrawals due to adverse effects (follow-up 4 weeks) 11 9/65 1/66 RR 9.14 (1.19 to 12 more per 100 (from 0 (13.8%) (1.5%) 70.10) more to 100 more) VERY LOW Primary outcome: constipation (adverse effects) (follow-up 4 weeks) 11 14/65 2/66 RR 7.11 (1.68 to 19 more per 100 (from 2 (21.5%) (3%) 30.04) more to 88 more) VERY LOW Primary outcome: somnolence/drowsiness (adverse effects) (follow-up 4 weeks) 11 8/65 4/66 RR 2.03 (0.64 to 6.42) 6 more per 100 (from 2 (12.3%) (6.1%) fewer to 33 more) VERY LOW Primary outcome: nausea (adverse effects) (follow-up 4 weeks) 11 15/65 2/66 RR 7.62 (1.81 to 20 more per 100 (from 2 (23.1%) (3%) 31.99) more to 94 more) VERY LOW Primary outcome: dizziness (adverse effects) (follow-up 4 weeks) 11 3/65 0/66 RR 7.11 (0.37 to – (4.6%) (0%) 134.91) VERY LOW Other reported pain outcome: pain intensity (Scale: VASpi-10 cm) (follow-up 4 weeks) 11 65 66 Treatment = 1.4 (0.1); Placebo = 2.2 (0.1) p < 0.001 LOW 1 Harati et al. (1998). 391

392 Note: no study on tramadol that reported the critical outcomes on pain was 393 identified or met the inclusion and exclusion criteria.

394 Table 23 Summary profile– oxycodone as monotherapy (placebo- 395 controlled trials) No of Relative risk Oxycodone Placebo Absolute risk Quality studies (95% CI) Primary outcome: withdrawals owing to adverse effects (follow-up 6 weeks) 11 7/82 4/77 RR 1.64 (0.50 to 5.39) 3 more per 100 (from 3 (8.5%) (5.2%) fewer to 23 more) VERY LOW Primary outcome: somnolence/drowsiness (adverse effects) (follow-up 6 weeks) 11 33/82 1/77 RR 30.99 (4.34 to 39 more per 100 (from 4 (40.2%) (1.3%) 221.09) more to 100 more) VERY LOW Primary outcome: nausea (adverse effects) (follow-up 6 weeks) 11 30/82 6/77 RR 4.70 (2.07 to 29 more per 100 (from 8 (36.6%) (7.8%) 10.65) more to 75 more) VERY LOW Primary outcome: dizziness (adverse effects) (follow-up 6 weeks) 11 26/82 8/77 RR 3.05 (1.47 to 6.33) 21 more per 100 (from 5 (31.7%) (10.4%) more to 55 more) VERY LOW Primary outcome: vomiting (adverse effects) (follow-up 6 weeks) 11 17/82 2/77 RR 7.98 (1.91 to 18 more per 100 (from 2 (20.7%) (2.6%) 33.41) more to 84 more) VERY LOW

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 39 of 150

Other reported pain outcome: pain intensity (Scale: NRSpi 11-point) (follow-up 6 weeks) 11 82 77 Treatment = −2.6 (2.54); Placebo = −1.5 (2.19) p < 0.001 LOW 1 Gimbel et al. (2003). 396

397 Note: no study on oxycodone that reported the critical outcomes on pain was 398 identified or met the inclusion and exclusion criteria.

399 Topical treatments 400 Table 24 Summary profile – topical capsaicin (0.075% cream) as 401 monotherapy (placebo-controlled trials) Topical No of capsaicin Placebo Relative risk Absolute risk Quality studies (0.075% cream (95% CI) cream) Primary outcome: patient-reported global improvement/impression of change (follow-up 8 weeks) 11 23/40 26/40 RR 0.88 (0.62 to 1.26) 8 fewer per 100 (from 25 (57.5%) (65%) fewer to 17 more) MODERATE Primary outcome: withdrawals owing to adverse effects (follow-up 8 weeks) 22 3/39 0/37 RR 3.84 (0.45 to - (7.7%) (0%) 32.92) LOW Primary outcome: burning (adverse effects) (follow-up 8 weeks) 22 23/39 7/37 RR 3.11 (1.52 to 6.37) 40 more per 100 (from 10 (59%) (18.9%) more to 100 more) LOW 1 Tandan et al. (1992). 2 Schefflet et al. (1991); Tandan et al. (1992). 402

403 Head-to-head comparative trials (monotherapy) 404 Table 25 Summary profile – pregabalin vs amitriptyline as monotherapy 405 (comparative trials) No of Relative risk Pregabalin Amitriptyline Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 50% pain reduction (follow-up 5 weeks) 11 21/51 15/51 RR 1.40 (0.82 to 2.39) 12 more per 100 (from 5 (41.2%) (29.4%) fewer to 41 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 5 weeks) 11 34/51 32/51 RR 1.08 (0.82 to 1.44) 5 more per 100 (from 11 (66.7%) (62.7%) fewer to 28 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 5 weeks) 11 6/51 17/51 RR 0.35 (0.15 to 0.82) 22 fewer per 100 (from 6 VERY (11.8%) (33.3%) fewer to 28 fewer) LOW Primary outcome: dizziness (adverse effects) (follow-up 5 weeks) 11 3/51 2/51 RR 1.5 (0.26 to 8.60) 2 more per 100 (from 3 VERY (5.9%) (3.9%) fewer to 30 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 5 weeks) 11 3/51 7/51 RR 0.43 (0.12 to 1.56) 8 fewer per 100 (from 12 VERY (5.9%) (13.7%) fewer to 8 more) LOW 1 Bansal et al. (2009). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 40 of 150

406

407 Table 26 Summary profile – amitriptyline vs gabapentin as monotherapy 408 (comparative trials) No of Relative risk Amitriptyline Gabapentin Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 11 14/21 11/21 RR 1.27 (0.77 to 2.11) 14 more per 100 (from 12 (66.7%) (52.4%) fewer to 58 more) MODERATE Primary outcome: number of withdrawals owing to adverse effects (follow-up 6 weeks) 11 2/25 1/25 RR 2.00 (0.19 to 4 more per 100 (from 3 VERY (8%) (4%) 20.67) fewer to 79 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 6 to 12 weeks) 22 28/37 22/38 RR 1.58 (0.49 to 5.15) 34 more per 100 (from 30 (75.7%) (57.9%) fewer to 100 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 12 weeks) 11 2/25 7/25 RR 0.29 (0.07 to 1.24) 20 fewer per 100 (from 26 VERY (8%) (28%) fewer to 7 more) LOW Primary outcome: sedation (adverse effects) (follow-up 12 weeks) 11 8/25 12/25 RR 0.67 (0.33 to 1.35) 16 fewer per 100 (from 32 VERY (32%) (48%) fewer to 17 more) LOW 1 Morello et al. (1999). 2 Morello et al. (1999); Dallocchio et al. (2000). 409

410 Table 27 Summary profile – amitriptyline vs topical capsaicin (0.075% 411 cream) as monotherapy (comparative trials) Capsaicin No of Relative risk Amitriptyline cream Absolute risk Quality studies (95% CI) (0.075%) Primary outcome: sedation (adverse effects) (follow-up 8 weeks) 11 69/117 0/118 ∞ (∞) – VERY (59%) (0%) LOW Primary outcome: burning (adverse effects) (follow-up 8 weeks) 11 0/117 68/118 0.00 (0.00, ∞) _ VERY (0%) (57.6%) LOW Other reported pain outcome: pain relief (Scale: VASpr-100 mm) (follow-up 8 weeks) 11 108 104 Amitriptyline = 57.0 (3.6); Capsaicin cream = 55.1 (3.5), p > 0.05 LOW Other reported pain outcome: pain intensity (Scale: VASpi-100 mm) (follow-up 8 weeks) 11 108 104 Amitriptyline = −29.1 (2.9); Capsaicin cream = −26.1 (2.9), p > 0.05 LOW 1 Biesbroeck et al. (1995). 412

413 Note: no study on amitriptyline vs topical capsaicin (0.075% cream) that 414 reported the critical outcomes on pain was identified or met the inclusion and 415 exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 41 of 150

416 Head-to-head comparative trials (combination therapy) 417 Table 28 Summary profile – gabapentin + oxycodone as combination 418 therapy vs gabapentin alone (comparative trials) No of Gabapentin + Gabapentin Relative risk Absolute risk Quality studies oxycodone alone (95% CI) Primary outcome: number of withdrawals owing to adverse effects (follow-up 12 weeks) 11 27/168 9/167 RR 3.00 (1.45 to 6.19) 11 more per 100 (from 2 VERY (16%) (5.3%) more to 28 more) LOW Primary outcome: constipation (adverse effects) (follow-up 12 weeks) 11 45/168 10/167 RR 4.47 (2.33 to 8.58) 21 more per 100 (from 8 VERY (26.8%) (6%) more to 45 more) LOW Primary outcome: nausea (adverse effects) (follow-up 12 weeks) 11 43/168 18/167 RR 2.37 (1.43 to 3.94) 15 more per 100 (from 5 VERY (25.6%) (10.8%) more to 32 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 12 weeks) 11 25/168 6/167 RR 4.14 (1.74 to 9.84) 11 more per 100 (from 3 VERY (14.9%) (3.6%) more to 32 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 12 weeks) 11 37/168 9/167 RR 4.09 (2.04 to 8.20) 17 more per 100 (from 6 VERY (22%) (5.4%) more to 39 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 12 weeks) 11 147/168 119/167 RR 1.23 (1.10 to 1.37) 16 more per 100 (from 7 (87.5%) (71.3%) more to 26 more) MODERATE Other non-primary outcome: pain relief (scale: box scale-11) (follow-up 12 weeks) 11 169 169 Gabapentin + Oxycodone = 2.1 (2.61); Gabapentin = 1.5 (2.38), p = 0.007 LOW 1 Hanna et al. (2008). 419

420 Note: no study on gabapentin plus oxycodone as combination therapy vs 421 gabapentin alone that reported the critical outcomes on pain was identified or 422 met the inclusion and exclusion criteria.

423 3.2.3 Evidence statements

424 For details of how the evidence is graded, see ‘The guidelines manual’.

425 3.2.3.1 No study on clomipramine, dosulepin (dothiepin), doxepin, 426 imipramine, lofepramine, nortriptyline, trimipramine, citalopram, 427 fluoxetine, paroxetine, sertraline, carbamazepine, phenytoin, 428 buprenorphine, co-codamol, codeine phosphate, co-dydramol, 429 dihydrocodeine, fentanyl, morphine and topical lidocaine was 430 identified or met the inclusion and exclusion criteria for PDN.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 42 of 150

431 Antidepressants as monotherapy against placebo 432 Desipramine (linked to table 13) 433 Critical outcomes (pain) 434 3.2.3.2 Moderate quality evidence from one study with 40 patients with 435 PDN, showed that desipramine is more effective than placebo in 436 achieving patient-reported global improvement/impression of 437 change from baseline up to 6 weeks’ follow-up.

438 Critical outcomes (adverse effects) 439 3.2.3.3 Very low quality evidence from one study with 40 patients with 440 PDN, showed that there is no significant difference between 441 desipramine and placebo in patients withdrawing from studies due 442 to adverse effects, dry mouth, sedation or any adverse effects 443 (unspecified) from baseline up to 6 weeks’ follow-up.

444 Duloxetine (linked to table 14) 445 Critical outcomes (pain) 446 3.2.3.4 Moderate quality evidence from two studies with 542 patients with 447 PDN, showed that there is no significant difference between 448 duloxetine and placebo in achieving at least 30% pain reduction 449 from baseline up to 12 weeks’ follow-up.

450 3.2.3.5 Moderate quality evidence from four studies with 1339 patients with 451 PDN, showed that duloxetine is more effective than placebo in 452 achieving at least 50% pain reduction.

453 Critical outcomes (adverse effects) 454 3.2.3.6 Moderate quality evidence from four studies with 1354 patients with 455 PDN and three studies with 1006 patients with PDN, showed that 456 patients on duloxetine are more likely to withdraw from studies due 457 to adverse effects and to experience dizziness compared with 458 placebo from baseline up to 12 weeks’ follow-up.

459 3.2.3.7 Low quality evidence from two studies with 672 patients with PDN, 460 showed that there is no significant difference between patients on

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 43 of 150

461 duloxetine and placebo in experiencing dry mouth from baseline up 462 to 12 weeks’ follow-up.

463 3.2.3.8 Low quality evidence from two studies with 549 patients with PDN, 464 showed that patients on duloxetine are more likely to experience 465 gastrointestinal disturbances compared with placebo from baseline 466 up to 12 weeks’ follow-up.

467 3.2.3.9 Very low quality evidence from one study with 215 patients with 468 PDN, showed that there is no significant difference between 469 patients on duloxetine and placebo in experiencing vomiting from 470 baseline up to 12 weeks’ follow-up.

471 3.2.3.10 Moderate quality evidence from one study with 215 patients with 472 PDN, showed that there is no significant difference between 473 patients on duloxetine and placebo in experiencing any adverse 474 effects (unspecified) from baseline up to 12 weeks’ follow-up.

475 Venlafaxine (linked to table 15) 476 Critical outcomes (pain) 477 3.2.3.11 Moderate quality evidence from one study with 243 patients with 478 PDN, showed that there is no significant difference between 479 venlafaxine and placebo in achieving at least 50% pain reduction 480 from baseline up to 6 weeks’ follow-up.

481 Critical outcomes (adverse effects) 482 3.2.3.12 Very low quality evidence from one study with 217 patients with 483 PDN, showed that there is no significant difference between 484 patients on venlafaxine and placebo in experiencing vomiting from 485 baseline up to 6 weeks’ follow-up.

486 Anti-epileptics as monotherapy against placebo 487 Gabapentin (linked to table 16) 488 Critical outcomes (pain) 489 3.2.3.13 Moderate quality evidence from two studies with 209 patients with 490 PDN, showed that gabapentin is more effective than placebo in

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 44 of 150

491 achieving patient-reported global improvement/impression of 492 change from baseline up to 8 weeks’ follow-up.

493 Critical outcomes (adverse effects) 494 3.2.3.14 Low quality evidence from two studies with 225 patients with PDN, 495 showed that there is no significant difference between patients on 496 gabapentin and placebo withdrawing from studies due to adverse 497 effects from baseline up to 8 weeks’ follow-up.

498 3.2.3.15 Low quality evidence from two studies with 219 patients with PDN 499 and also another two studies with 219 patients with PDN, showed 500 that patients on gabapentin are more likely to experience dizziness 501 and somnolence compared with placebo from baseline up to 502 8 weeks’ follow-up.

503 Pregabalin (linked to table 17) 504 Critical outcomes (pain) 505 3.2.3.16 Moderate quality evidence from one study with 259 patients, and 506 another five studies with 1267 patients with PDN, showed that 507 pregabalin is more effective than placebo in achieving at least 30% 508 and at least 50% pain reduction from baseline up to 14 weeks’ 509 follow-up.

510 3.2.3.17 Low quality evidence from two studies with 570 patients with PDN, 511 showed that pregabalin is more effective than placebo in achieving 512 patient-reported global improvement/impression of change from 513 baseline up to 12 weeks’ follow-up.

514 Critical outcomes (adverse effects) 515 3.2.3.18 Moderate quality evidence from six studies with 1608 patients with 516 PDN, showed that more patients on pregabalin withdraw from 517 studies due to adverse effects, or experience dizziness and 518 somnolence compared with placebo from baseline up to 14 weeks’ 519 follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 45 of 150

520 3.2.3.19 Low quality evidence from four studies with 1125 patients with 521 PDN, showed that patients on pregabalin are more likely to 522 experience weight gain compared with placebo from baseline up to 523 14 weeks’ follow-up.

524 3.2.3.20 Moderate quality evidence from two studies with 463 patients with 525 PDN, showed that patients on pregabalin are more likely to 526 experience any adverse effects (unspecified) compared with 527 placebo from baseline up to 14 weeks’ follow-up.

528 Lamotrigine (linked to table 18) 529 Critical outcomes (pain) 530 3.2.3.21 Moderate quality evidence from two studies with 444 patients, three 531 studies with 497 patients, and one study with 43 patients with PDN, 532 showed that there is no significant difference between lamotrigine 533 and placebo in achieving at least 30% or at least 50% pain 534 reduction and global improvement/impression of change 535 respectively from baseline up to 19 weeks’ follow-up.

536 Critical outcomes (adverse effects) 537 3.2.3.22 Low quality evidence from four studies with 799 patients, and three 538 studies with 759 patients with PDN, showed that there is no 539 significant difference between patients on lamotrigine and placebo 540 in withdrawal due to adverse effects and dizziness from baseline up 541 to 19 weeks’ follow-up.

542 3.2.3.23 High quality evidence from three studies with 759 patients with 543 PDN, showed that there is no significant difference between 544 patients on lamotrigine and placebo in experiencing any adverse 545 effects (unspecified) from baseline up to 19 weeks’ follow-up.

546 Topiramate (linked to table 19) 547 Critical outcomes (pain) 548 3.2.3.24 Moderate quality evidence from one study with 317 patients with 549 PDN, showed that topiramate is more effective than placebo in 550 achieving at least 30% or at least 50% pain reduction and global Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 46 of 150

551 improvement/impression of change from baseline up to 12 weeks’ 552 follow-up.

553 Critical outcomes (adverse effects) 554 3.2.3.25 High quality evidence from two studies with 1592 patients with 555 PDN, showed that patients on topiramate are more likely to 556 withdraw from studies due to adverse effects compared with 557 placebo from baseline up to 22 weeks’ follow-up.

558 3.2.3.26 Moderate quality evidence from two studies with 1589 patients with 559 PDN, showed that patients on topiramate are more likely to 560 experience somnolence and fatigue compared with placebo from 561 baseline up to 22 weeks’ follow-up.

562 3.2.3.27 Very low quality evidence from one study with 320 patients with 563 PDN, showed that there is no significant difference between 564 patients on topiramate and placebo in experiencing dizziness from 565 baseline up to 22 weeks’ follow-up.

566 3.2.3.28 Moderate quality evidence from one study with 323 patients with 567 PDN, showed that there is no significant difference between 568 patients on topiramate and placebo in experiencing any adverse 569 effects (unspecified) from baseline up to 22 weeks’ follow-up.

570 Oxcarbazepine (linked to table 20) 571 Critical outcomes (pain) 572 3.2.3.29 Moderate quality evidence from one study with 146 patients with 573 PDN, showed that oxcarbazepine is more effective than placebo in 574 achieving at least 30% and at least 50% pain reduction from 575 baseline up to 16 weeks’ follow-up.

576 3.2.3.30 Moderate quality evidence from two studies with 378 patients with 577 PDN, showed that there is no significant difference between 578 oxcarbazepine and placebo in achieving global 579 improvement/impression of change from baseline up to 16 weeks.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 47 of 150

580 Critical outcomes (adverse effects) 581 3.2.3.31 Moderate quality evidence from three studies with 634 patients with 582 PDN, showed that patients on oxcarbazepine are more likely to 583 withdraw from studies due to adverse effects compared with 584 placebo from baseline up to 16 weeks’ follow-up.

585 3.2.3.32 Low quality evidence from two studies with 469 patients with PDN, 586 showed that patients on oxcarbazepine are more likely to 587 experience dizziness and somnolence compared with placebo from 588 baseline up to 16 weeks’ follow-up.

589 3.2.3.33 Low quality evidence from two studies with 469 patients with PDN, 590 showed that there is no significant difference between patients on 591 oxcarbazepine and placebo in experiencing fatigue from baseline 592 up to 16 weeks’ follow-up.

593 Sodium valproate (linked to table 21) 594 Critical outcomes (pain) 595 3.2.3.34 No study on sodium valproate that reported the critical outcomes 596 on pain was identified or met the inclusion and exclusion criteria.

597 Critical outcomes (adverse effects) 598 3.2.3.35 Low quality evidence from two studies with 103 patients with PDN, 599 showed that there is no significant difference between patients on 600 sodium valproate and placebo withdrawing from studies due to 601 adverse effects from baseline up to 12 weeks’ follow-up.

602 3.2.3.36 Very low quality evidence from one study with 40 patients with 603 PDN, showed that there is no significant difference between 604 patients on sodium valproate and placebo in experiencing any 605 adverse effects (unspecified) from baseline up to 12 weeks’ follow- 606 up.

607 Other reported pain outcomes 608 3.2.3.37 Low quality evidence from one study with 40 patients with PDN, 609 showed that there is no significant difference on pain intensity

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 48 of 150

610 scores between patients on sodium valproate and placebo from 611 baseline up to 12 weeks’ follow-up.

612 3.2.3.38 Low quality evidence from one study with 43 patients with PDN, 613 showed that patients on sodium valproate are more likely to have 614 better scores in pain relief scale than placebo from baseline up to 615 12 weeks’ follow-up.

616 Opioid analgesics as monotherapy against placebo 617 Tramadol (linked to table 22) 618 Critical outcomes (pain) 619 3.2.3.39 No study on tramadol that reported the critical outcomes on pain 620 was identified or met the inclusion and exclusion criteria.

621 Critical outcomes (adverse effects) 622 3.2.3.40 Very low quality evidence from one study with 131 patients with 623 PDN, showed that patients on tramadol are more likely to withdraw 624 from studies due to adverse effects, constipation and nausea 625 compared with placebo from baseline up to 4 weeks’ follow-up.

626 3.2.3.41 Very low quality evidence from one study with 131 patients with 627 PDN, showed that there is no significant difference between 628 patients on tramadol and placebo in experiencing somnolence and 629 dizziness from baseline up to 4 weeks’ follow-up.

630 Other reported pain outcomes 631 3.2.3.42 Low quality evidence from one study with 131 patients with PDN, 632 showed that patients on tramadol are more likely to have better 633 scores in pain intensity scale than placebo from baseline up to 634 4 weeks’ follow-up.

635 Oxycodone (linked to table 23) 636 Critical outcomes (pain) 637 3.2.3.43 No study on oxycodone that reported the critical outcomes on pain 638 was identified or met the inclusion and exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 49 of 150

639 Critical outcomes (adverse effects) 640 3.2.3.44 Very low quality evidence from one study with 159 patients with 641 PDN, showed that patients on oxycodone are more likely to 642 experience somnolence, nausea, dizziness and vomiting compared 643 with placebo from baseline up to 6 weeks’ follow-up.

644 3.2.3.45 Very low quality evidence from one study with 159 patients with 645 PDN, showed that there is no significant difference between 646 patients on oxycodone and placebo withdrawing from studies due 647 to adverse effects from baseline up to 6 weeks’ follow-up.

648 Other reported pain outcomes 649 3.2.3.46 Low quality evidence from one study with 159 patients with PDN, 650 showed that patients on oxycodone are more likely to have better 651 scores in pain intensity scale than placebo from baseline up to 652 6 weeks’ follow-up.

653 Topical treatments as monotherapy against placebo 654 Topical capsaicin (0.075% cream) (linked to table 24) 655 Critical outcomes (pain) 656 3.2.3.47 Moderate quality evidence from one study with 80 patients with 657 PDN, showed that there is no significant difference between 658 patients on topical capsaicin (0.075% cream) and placebo in 659 achieving global improvement/impression of change from baseline 660 up to 8 weeks’ follow-up.

661 Critical outcomes (adverse effects) 662 3.2.3.48 Low quality evidence from two studies with 76 patients with PDN, 663 showed that patients on topical capsaicin (0.075% cream) are more 664 likely to experience burning compared with placebo from baseline 665 up to 8 weeks’ follow-up.

666 3.2.3.49 Low quality evidence from two studies with 76 patients with PDN, 667 showed that there is no significant difference between patients on 668 topical capsaicin (0.075% cream) and placebo withdrawing from

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 50 of 150

669 studies due to adverse effects from baseline up to 8 weeks’ follow- 670 up.

671 Head-to-head comparative trials (monotherapy) 672 Pregabalin vs amitriptyline (linked to table 25) 673 Critical outcomes (pain) 674 3.2.3.50 Moderate quality evidence from one study with 102 patients with 675 PDN, showed that there is no significant difference between 676 patients on pregabalin and patients on amitriptyline in achieving at 677 least 50% pain reduction and global improvement/impression of 678 change from baseline up to 5 weeks’ follow-up.

679 Critical outcomes (adverse effects) 680 3.2.3.51 Very low quality evidence from one study with 102 patients with 681 PDN, showed that there is no significant difference between 682 patients on pregabalin and patients on amitriptyline withdrawing 683 from studies due to adverse effects, or experiencing dizziness and 684 somnolence from baseline up to 5 weeks’ follow-up.

685 Amitriptyline vs gabapentin (linked to table 26) 686 Critical outcomes (pain) 687 3.2.3.52 Moderate quality evidence from one study with 42 patients with 688 PDN, showed that there is no significant difference between 689 patients on amitriptyline and patients on gabapentin in achieving 690 global improvement/impression of change from baseline up to 6 691 weeks’ follow-up.

692 Critical outcomes (adverse effects) 693 3.2.3.53 Very low quality evidence from one study with 50 patients with 694 PDN, showed that there is no significant difference between 695 patients on amitriptyline and patients on gabapentin withdrawing 696 from studies due to adverse effects, or experiencing dizziness and 697 sedation from baseline up to 12 weeks’ follow-up.

698 3.2.3.54 Low quality evidence from two studies with 75 patients with PDN, 699 showed that there is no significant difference between patients on Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 51 of 150

700 amitriptyline and patients on gabapentin in experiencing any 701 adverse effects (unspecified) from baseline up to 12 weeks’ follow- 702 up.

703 Amitriptyline vs topical capsaicin (0.075% cream) (linked to table 27) 704 Critical outcomes (pain) 705 3.2.3.55 No study on amitriptyline vs topical capsaicin (0.075% cream) that 706 reported the critical outcomes on pain was identified or met the 707 inclusion and exclusion criteria.

708 Critical outcomes (adverse effects) 709 3.2.3.56 Very low quality evidence from one study with 235 patients with 710 PDN, showed that there is no significant difference between 711 patients on amitriptyline and patients on topical capsaicin (0.075% 712 cream) in experiencing sedation and burning from baseline up to 8 713 weeks’ follow-up.

714 Other reported pain outcomes 715 3.2.3.57 Low quality evidence from one study with 212 patients with PDN, 716 showed that there is no significant difference on pain intensity 717 scores and pain relief scores between patients on amitriptyline and 718 patients on topical capsaicin (0.075% cream) from baseline up to 8 719 weeks’ follow-up.

720 Head-to-head comparative trial (combination therapy) 721 Gabapentin + oxycodone as combination therapy vs gabapentin alone 722 (linked to table 28) 723 Critical outcomes (pain) 724 3.2.3.58 No study on gabapentin + oxycodone as combination therapy vs 725 gabapentin alone that reported the critical outcomes on pain was 726 identified or met the inclusion and exclusion criteria.

727 Critical outcomes (adverse effects) 728 3.2.3.59 Very low quality evidence from one study with 335 patients with 729 PDN, showed that there patients on gabapentin + oxycodone are 730 more likely to withdraw from studies due to adverse effects, or

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 52 of 150

731 experience constipation, nausea, dizziness and somnolence 732 compared with gabapentin alone from baseline up to 12 weeks’ 733 follow-up.

734 3.2.3.60 Moderate quality evidence from one study with 335 patients with 735 PDN, showed that patients on gabapentin + oxycodone are more 736 likely to experience any adverse effects (unspecified) compared 737 with gabapentin alone from baseline up to 12 weeks’ follow-up.

738 Other reported pain outcomes 739 3.2.3.61 Low quality evidence from one study with 338 patients with PDN, 740 showed that patients on gabapentin + oxycodone are more likely to 741 have better scores in pain relief scale than gabapentin alone from 742 baseline up to 12 weeks’ follow-up.

743 3.2.4 Health economic modelling

744 This is a summary of the modelling carried out for this review question. See 745 appendix 10.11 for full details of the modelling carried out for the guideline.

746 The analysis presented results in terms of decreasing mean net monetary 747 benefit (NMB) associated with each drug at a threshold of £20,000 and 748 £30,000 per QALY gained. All comparisons were made with placebo.

749 The cost effectiveness results for PDN are presented in table 1HE and table 750 2Table HE. It was not possible to calculate an incremental based on the 751 results presented in the draft report due to rounding. The mean net benefits 752 indicated that the two most cost effective treatments are duloxetine 60mg and 753 20mg. All the other treatments were associated with higher costs and lower 754 effectiveness. At £20,000 threshold only duloxetine 60mg, 20mg, gabapentin 755 3600mg and venlafaxine were cost effective treatment options. At £30,000 756 duloxetine 120mg and pregabalin (300-600mg) became cost effective as well. 757 At both thresholds, treatments associated with the highest probability of being 758 cost effective were duloxetine 60mg, 20mg and gabapentin 3600mg. It was 759 noted from standard deviations around the point estimates of total QALYs and 760 costs that there was considerable overlap in terms of QALY gains, but that

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 53 of 150

761 there was significant difference in terms of cost. This suggested that the 762 acquisition cost of the treatments was the main driver of differences between 763 the treatment options.

764

765 Table 29 PDN incremental cost effectiveness results Mean incremental net benefit (£) per Drug person at a threshold per QALY of: £30,000 £20,000 Single dose comparators Duloxetine 60 mg 4375 2311 Duloxetine 20 mg 4088 2376 Gabapentin 3600 mg 3632 2057 Venlafaxine 225 mg 2798 1730 Duloxetine 120 mg 1192 −577 Pregabalin 600 mg 665 −1180 Pregabalin 300 mg 386 −1351 Venlafaxine 75 mg −141 −159 Oxcarbazepine 1200 mg −2783 −2800 Oxcarbazepine 600 mg −3133 −2568 Pregabalin 150 mg −3530 −3729 Topiramate 400 mg −3903 −5190 Oxcarbazepine 1800 mg −6119 −5368 Single and flexible dose comparators Duloxetine 60 mg 4375 2311 Duloxetine 20 mg 4088 2376 Gabapentin (3600 mg) 3632 2057 Venlafaxine 225 mg 2798 1730 Duloxetine 120 mg 1192 −577 Pregabalin flexible dose (150–600 mg) −126 −1665 Venlafaxine 75 mg -141 −159 Topiramate 400 mg −3903 −5190 Oxcarbazepine flexible dose (600– 1800 mg) −4941 −4281 766

767 Table 30 PDN probabilistic results Single dose analysis Flexible dose analysis Probability of being the most Probability of being the most cost-effective drug at a cost-effective drug at a Drug threshold per QALY of: Drug threshold per QALY of: £30,000 £20,000 £30,000 £20,000 Duloxetine Duloxetine 0.321 0.301 0.348 0.307 60 mg 60 mg Duloxetine Duloxetine 0.293 0.328 0.312 0.333 20 mg 20 mg Gabapentin 0.191 0.218 Gabapentin 0.210 0.222

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 54 of 150

3600 mg (3600 mg) Pregabalin Venlafaxine 0.073 0.017 0.081 0.124 300 mg 225 mg Venlafaxine Duloxetine 0.070 0.121 0.047 0.009 225 mg 120 mg Duloxetine Venlafaxine 0.042 0.009 0.001 0.005 120 mg 75 mg Pregabalin Pregabalin 0.009 0.001 flexible dose 0.001 0.000 600 mg (150–600 mg) Venlafaxine Topiramate 0.001 0.005 0.000 0.000 75 mg 400 mg Oxcarbazepine Oxcarbazepine 0.000 0.000 flexible dose 0.000 0.000 1200 mg (600–1800 mg) Oxcarbazepine 0.000 0.000 Placebo 0.000 0.000 600 mg Oxcarbazepine 0.000 0.000 1800 mg Pregabalin 0.000 0.000 150 mg Topiramate 0.000 0.000 400 mg Placebo 0.000 0.000 768

769 Sensitivity analysis and uncertainty 770 Numerous sensitivity analyses were conducted to explore how the model’s 771 inputs affected its results and, in particular, the extent to which single 772 parameters would need to be altered before different options became cost 773 effective (‘threshold analysis’).. At a £30,000 threshold duloxetine 60 mg was 774 the most cost-effective option across the analyses apart from:

775 When key clinical parameters were equalised across the treatments, 776 duloxetine 20 mg became the most cost-effective option. 777 If gabapentin 3600 mg was free; it became the most cost-effective option. 778 Although the differences in NMB between gabapentin 3600 mg and 779 duloxetine 60 mg was still very small. 780 When duloxetine 60 mg, duloxetine 20 mg and gabapentin 3600 mg all had 781 very high and similar NMB and all have high and occasionally similar 782 probabilities of being cost effective.

783 All the duloxetine doses, gabapentin 3600 mg and pregabalin 300 mg were all 784 associated with positive NMB across all scenarios. The highest dose for 785 oxcarbazepine is never cost effective and topiramate is only cost effective if its 786 price is reduced to zero. Venlafaxine 75 mg became the most cost-effective

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 55 of 150

787 option when all clinical parameters were equalised across treatments because 788 it was the least costly.

789 At a £20,000 threshold, the results of the sensitivity analysis changed slightly. 790 Under most scenarios duloxetine 20 mg was the most cost-effective option. 791 There was greater variation in the results with alternatives having higher 792 NMBs. However, for the majority of the analyses the results mirrored those 793 obtained at a £30,000 threshold, including the results of the probabilistic 794 analysis.

795 Under the flexible dose analysis the results are very similar to the single-dose 796 analysis. The price of pregabalin has to fall to 80% and 20% of its current 797 value for it to be considered cost effective at £30,000 and £20,000 thresholds 798 respectively.

799 Health economics evidence statements – PDN 800 3.2.4.1 Partially applicable evidence from one study with minor limitations, 801 showed that duloxetine was the most cost-effective treatment for 802 PDN compared with gabapentin, oxcarbazepine, pregabalin, 803 topiramate and venlafaxine

804 3.2.4.2 Partially applicable evidence from one study with minor limitations, 805 showed that gabapentin was the second most cost-effective 806 treatment for PDN compared with oxcarbazepine, pregabalin, 807 topiramate and venlafaxine

808 3.2.5 Evidence to recommendations

Relative value of The GDG agreed and endorsed the international IMMPACT different recommendations that the critical outcomes on pain should be both outcomes subjective measures of pain reduction (patient-reported at least 30% pain reduction and at least 50% pain reduction on a numerical rating scale or visual analogue scale), and the overall feeling of well-being reported by the patients (patient-reported global improvement/impression of change). The GDG agreed that comparing mean scores from a 10-point pain scale between groups was of less importance because it is more prone to bias and does not illustrate the proportion of patients achieving a certain

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 56 of 150

magnitude of effects.

The GDG also agreed that, apart from the efficacy of pharmacological treatments on pain outcomes, the adverse effects of individual treatments should also be considered to balance the benefit and harm to patients. Because numerous specific adverse effects are related to each pharmacological treatment, the GDG decided to define between five and seven critical adverse effects for each class of drug that they would consider when making recommendations. A questionnaire was completed by GDG members to select the critical adverse effects outcomes (see appendix 10.3A for the questionnaire and selected outcomes).

Quality of The GDG agreed that when discussing the quality of evidence, evidence consideration of the number of studies, the size of the study population and the magnitude of effects are important.

Overall, the GDG agreed that the core evidence-base is from placebo- controlled trials, and evidence on head-to-head comparative trials and trials on combination therapy is very limited. Hence, the GDG felt that they could not confidently draw conclusions solely based on this evidence. The focus of the discussion was based on the placebo-controlled trials and evidence from health economics evaluation.

Placebo-controlled trials For antidepressants, only studies on duloxetine (an SNRI), venlafaxine (an SNRI) and desipramine (a TCA) were identified or met the inclusion and exclusion criteria for the PDN analysis. The GDG agreed that all three drugs have moderate-quality evidence on the critical pain outcomes but most evidence was for duloxetine.

Hence, the GDG agreed that duloxetine seems to have better evidence of efficacy compared with venlafaxine and desipramine.

For anti-epileptics, the GDG agreed that evidence on the efficacy of lamotrigine and sodium valproate was insufficient. The GDG also agreed that evidence for the efficacy of topiramate and oxcarbazepine was limited and most evidence seems to be for pregabalin and gabapentin (with moderate-quality). Therefore, the GDG felt that pregabalin and gabapentin should be the focus for discussion.

For opioid analgesics and topical treatment, the GDG agreed that evidence on the efficacy of topical capsaicin (0.075% cream) was insufficient, and the Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 57 of 150

low-quality evidence on tramadol and oxycodone was for non-critical outcomes.

Trade-off between Desipramine clinical benefits Although there was some evidence for the efficacy of desipramine, it is no and harms longer listed in the BNF, and so should not be used in clinical practice.

Venlafaxine Based on information from the MHRA, the GDG agreed that the use of venlafaxine would need specialist care and regular monitoring, and so it should not be initiated in non-specialist settings.

Topiramate and oxcarbazepine The evidence showed that patients on either of these drugs were more likely to withdraw because of adverse effects than patients on gabapentin or pregabalin.

Duloxetine, gabapentin and pregabalin Duloxetine Cost-effectiveness evidence (see section below on economic considerations) demonstrated that duloxetine was the most cost-effective treatment for painful diabetic neuropathy (PDN). Therefore, the GDG decided that duloxetine should be recommended as first-line treatment for people with PDN. The GDG also agreed that the adverse effects of duloxetine, as well as the special warnings and precautions for its use as specified in the SPC (based on MHRA advice), should be discussed with the person and weighed against the benefit provided.

If duloxetine is contraindicated, the GDG agreed that amitriptyline should be the alternative antidepressant based on evidence from head-to-head comparative trials (see evidence statements 3.2.3.50 and 3.2.3.52), which indicated that amitriptyline is equally as effective as gabapentin and pregabalin for PDN.

Furthermore, because amitriptyline (a TCA) has different pharmacological profiles compared with duloxetine (an SNRI), the GDG agreed that amitriptyline also has a role as second-line treatment if patients did not have satisfactory pain reduction on duloxetine, based on the extrapolation of evidence from PHN and other neuropathic pain conditions.

Although the GDG agreed with the role of amitriptyline, they were also

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 58 of 150

concerned that many people who have satisfactory pain reduction with amitriptyline as first-line or second-line treatment would not be able to tolerate its adverse effects. The GDG reached a consensus that, in these cases other TCAs, namely nortriptyline and imipramine, should be recommended as alternatives to amitriptyline, because there is extrapolated evidence (see section 3.4.7) on efficacy in relation to global improvement for these drugs. Both are relatively low-cost drugs, and for this patient population they are potentially cost effective, provided that they do not cause other adverse effects that would reduce the potential gain in quality of life obtained by switching from amitriptyline.

Gabapentin and pregabalin Because pregabalin and gabapentin have similar pharmacological profiles (that is, both have high affinity for the alpha-2-delta subunit of the voltage- dependent calcium channel in the central nervous system – therefore if a person had unsatisfactory pain reduction with one drug, it is highly unlikely that they would achieve pain reduction with the other), and the cost- effectiveness evidence (see section below on economic considerations) demonstrated that gabapentin was more cost-effective than pregabalin for PDN, the GDG agreed that gabapentin should be the second-line treatment for PDN as monotherapy or as combination therapy with duloxetine (or amitriptyline).

Because gabapentin and pregabalin have similar efficacy, the GDG also agreed that pregabalin should be an alternative if gabapentin is effective but the person cannot tolerate the adverse effects or has difficulty adhering to the dosage schedule.

Economic The evidence from the cost effectiveness analysis indicated that duloxetine considerations was the most cost-effective treatment for most of the doses explored in the analysis and therefore the GDG recommended it as first-line treatment.

The GDG noted that no cost effectiveness evidence was presented on the role of amitriptyline, as the draft HTA report searches only extended to 2009 and only included placebo controlled trials, therefore, did not include head- to-head trials of amitriptyline and gabapentin.. However, the GDG noted that there was evidence from the clinical review (see clinical review) and their own clinical experience that indicated amitriptyline was as effective as pregabalin and gabapentin. The GDG considered that any differences in the rates of adverse events that did not lead to discontinuation would have an insignificant impact on the cost effectiveness results. The GDG was mindful Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 59 of 150

that the results from the cost effectiveness analysis were driven by the acquisition price of the treatments. The GDG noted that amitriptyline’s acquisition price (approximately £4 per 6 week treatment at 75mg at 2011 prices from the NHS drug tariff) was significantly lower than gabapentin’s (£13.91 per 6 week treatment at 1800mg at 2011 prices from teh NHS drug tariff) and pregablin’s (£96.60 per 6 weeks of treatment at 2011 prices from teh NHS drug tariff). Therefore, the GDG concluded that that amitriptyline represented a cost effective alternative to duloxetine.

The GDG considered that it was not appropriate to use the results of the draft HTA report to examine sequencing of treatments as the model did not consider class effects, titration practices and treatment switching. These factors resulted in sequences based solely on the outcome of the economic model being clinically inappropriate. The GDG considered the relative cost effectiveness of gabapentin and pregabalin. It acknowledged that gabapentin was most likely to be cost effective because of its lower acquisition cost. However, if gabapentin was effective but the person could not tolerate the adverse events then pregabalin represented a cost-effective alternative.

Other The GDG agreed that if first-line treatment did not result in satisfactory pain considerations reduction, a drug from another therapeutic class should be recommended as second-line treatment, either as monotherapy or as combination therapy with first-line treatment, instead of trying another drug from the same therapeutic class.

The GDG also agreed that if first-line and second-line treatment did not result in satisfactory pain reduction, the person should be referred to a specialist pain service and/or a condition-specific service.

Although evidence on tramadol and oxycodone was of low-quality and investigated non-critical outcomes, the GDG felt that opioid analgesics could be recommended as third-line treatment as rescue analgesics to ensure the continuity of treatment while a person is waiting for referral to a specialist pain service and/or a condition-specific service.

Because the GDG was concerned about the risk of long-term dependence, the severe adverse effects and the potential fatality of overdose with oxycodone, the GDG felt oxycodone should not be initiated without an assessment by a specialist pain service or a condition-specific service. However, the GDG also came to the consensus that recommending

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 60 of 150

tramadol was valid and appropriate as third-line treatment for neuropathic pain in non-specialist settings, either as monotherapy or as combination therapy with second-line treatment, because this drug is already commonly used in non-specialist settings.

The GDG agreed that there is a lack of evidence (especially placebo- controlled trials) for the efficacy of topical lidocaine for treating neuropathic pain in non-specialist settings. However, based on the clinical experience of members, the GDG acknowledged that a subgroup of people with ‘localised neuropathic pain’ who are unable to take oral medication because of medical conditions and/or disability may benefit from topical lidocaine. In view of the lack of evidence for PDN, the GDG felt that it could not recommend the use of topical lidocaine as first-line or second-line treatment. However, topical lidocaine may have a role as a rescue analgesic (while waiting for a referral to a specialist pain service) in a very small subgroup of people with localised pain who are unable to take oral medication because of medical conditions and/or disability.

Because amitriptyline is not licensed for neuropathic pain, the GDG came to the consensus that its initial dosage and titration should be lower than as recommended for this indication in the BNF.

809

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 61 of 150

810 3.2.6 Recommendations and research recommendations for 811 PDN

812 Recommendations

First-line treatment 1.1.10 Offer oral amitriptyline* or gabapentin as first-line treatment (see recommendation 1.1.13 for people with painful diabetic neuropathy).

1.1.11 For people with painful diabetic neuropathy, offer oral duloxetine as first-line treatment. If duloxetine is contraindicated, offer oral amitriptyline*.

1.1.12 Based on both the early and regular clinical reviews:

If improvement is satisfactory, continue the treatment; consider gradually reducing the dose over time if improvement is sustained. If amitriptyline* results in satisfactory pain reduction as first-line treatment but the person cannot tolerate the adverse effects, consider oral imipramine* or nortriptyline* as an alternative. If gabapentin results in satisfactory pain reduction as first-line treatment but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects, consider oral pregabalin as an alternative.

Second-line treatment 1.1.13 If satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person (see recommendation 1.1.16 for people

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 62 of 150

with painful diabetic neuropathy):

If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with oral gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with gabapentin (or pregabalin) switch to or combine with oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects).

1.1.14 For people with painful diabetic neuropathy, if satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person:

If first-line treatment was with duloxetine, switch to oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects) or switch to or combine with oral gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects).

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 63 of 150

Third-line treatment 1.1.15 If satisfactory pain reduction is not achieved with second-line treatment:

refer the person to a specialist pain service and/or a condition- specific service10 and while waiting for referral: consider oral tramadol as third-line treatment instead of or in combination11 with the second-line treatment consider a topical lidocaine patch for treatment of localised pain for people who are unable to take oral medication because of medical conditions and/or disability.

Other treatments 1.1.16 Do not start treatment with a topical capsaicin 8% patch or with opioids (such as morphine or oxycodone) other than tramadol without an assessment by a specialist pain service or a condition- specific service10.

1.1.17 Pharmacological treatments other than those recommended in this guideline that are started by a specialist pain service or a condition- specific service10 may continue to be prescribed in non-specialist settings, with a multidisciplinary care plan, local shared care agreements and careful management of adverse effects.

813

814 Research recommendations 815 See appendix B for full details of research recommendations.

10 A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. 11 The combination of tramadol with amitriptyline, nortriptyline, imipramine or duloxetine is associated with a low risk of serotonin syndrome (the features of which include confusion, delirium, shivering, sweating, changes in blood pressure and myoclonus). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 64 of 150

816 3.3 Post-herpetic neuralgia (PHN)

817 3.3.1 Review questions

818 Review question 1 819 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 820 analgesics, topical lidocaine and topical capsaicin as monotherapy (against 821 placebo) for the management of neuropathic pain condition (post-herpetic 822 neuralgia) in adults in non-specialist settings?

823 Review question 2 824 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 825 analgesics, topical lidocaine and topical capsaicin as combination therapy 826 (against monotherapy or other combination therapy) for the management of 827 neuropathic pain (post-herpetic neuralgia) in adults in non-specialist settings?

828 Review question 3 829 What is the clinical efficacy of antidepressants, compared with anti-epileptics, 830 opioid analgesics, topical lidocaine and topical capsaicin (and vice-versa) as 831 monotherapy for the management of neuropathic pain (post-herpetic 832 neuralgia) in adults in non-specialist settings?

833 3.3.2 Evidence review

834 A total of 21 randomised controlled trials were included for post-herpetic 835 neuralgia (PHN). Of the 34 included pharmacological treatments in (see table 836 4), no study was identified or met the inclusion and exclusion criteria for the 837 following pharmacological treatments (see table 31).

838 For the characteristics of included studies please see tables 32–36.

839 Table 31 Pharmacological treatments for which no study was identified 840 or met the inclusion and exclusion criteria for PHN Drug class: subclass Drug Antidepressants: tricyclic antidepressants (TCAs) Clomipramine Dosulepin (dothiepin) Doxepin Imipramine Lofepramine Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 65 of 150

Trimipramine Antidepressants: selective serotonin reuptake Citalopram inhibitors (SSRIs) Fluoxetine Paroxetine Sertraline Antidepressants: serotonin–norepinephrine reuptake Duloxetine inhibitors (SNRIs) Venlafaxine Anti-epileptics (anticonvulsants) Carbamazepine Lamotrigine Oxcarbazepine Phenytoin Sodium valproate Topiramate Opioid analgesics Buprenorphine Co-codamol Codeine phosphate Co-dydramol Dihydrocodeine Fentanyl Morphine Oxycodone 841

842 Table 32 Characteristics of included studies: antidepressants (placebo- 843 controlled trials)

Author Study Condition Treatment Titration or Mean Outcomes period (oral) fixed dosage dose (mg/day) (mg/day) Bowsher (1997) 3 months PHN Amitriptyline 25 NR N/A Graff-Radford et 8 weeks PHN Amitriptyline 12.5–200 NR Mean pain al. (2000) intensity score, AEs Max et al. (1988) 6 weeks PHN Amitriptyline 12.5–150 65 Global, AEs Kishore-Kumar et 6 weeks PHN Desipramine 12.5–250 167 Global, AEs al. (1990) PHN = post-herpetic neuralgia; Global = patient-reported global improvement; AEs = adverse effects. 844

845 Table 33 Characteristics of included studies: anti-epileptics (placebo- 846 controlled trials)

Author Study Condition Treatment (oral) Titration or Outcomes period fixed dosage (mg/day) Rice and Maton 7 weeks PHN Gabapentin 1800, 2400 50%, Global, (2001) AEs Rowbotham et 8 weeks PHN Gabapentin to 3600 Global, AEs al. (1998) Dworkin et 8 weeks PHN Pregabalin 150–600 30%, 50%, AEs Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 66 of 150

al.(2003) Sabatowski et al. 8 weeks PHN Pregabalin 150, 300 50%, Global, (2004) AEs Stacey et al. 4 weeks PHN Pregabalin 150–600, 600 30%, 50%, AEs (2008) van Seventer et 13 weeks PHN Pregabalin 150, 300, 600 30%, 50%, al. (2006) Global, AEs PHN = post-herpetic neuralgia; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 847

848 Table 34 Characteristics of included studies: opioid analgesics (placebo- 849 controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (oral) dosage (mg/day) Boureau et al. 6 weeks PHN Tramadol 100–400 50% (2003) PHN = post-herpetic neuralgia; 50% = at least 50% pain reduction 850

851 Table 35 Characteristics of included studies: topical capsaicin and 852 topical lidocaine (placebo-controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (topical) dosage (times/day) Bernstein et 6 weeks PHN Capsaicin 0.075% cream, 3 to 4 40%, AEs al. (1989) Watson et al. 6 weeks PHN Capsaicin 0.075% cream, 4 Mean change in (1993) pain relief score, AEs Backonja et 12 PHN Capsaicin 8% patch, applied 1-hour 30%, Global, AEs al. (2008) weeks with 3-months effects Irving et al. 12 PHN Capsaicin 8% patch, applied 1-hour 30%, 50%, Global, (2011) weeks with 3-months effects AEs Webster et al. 12 PHN Capsaicin 8% patch, applied 1-hour 30%, 50%, Global, (2010a) weeks with 3-months effects AEs Webster et al. 8 weeks PHN Capsaicin 8% patch, applied 1-hour 30%, 50%, Global, (2010b) with 3-months effects AEs Galer et al. 3 weeks PHN Lidocaine 5% patch, 1 Mean change in (2002) pain relief score PHN = post-herpetic neuralgia; PDN = painful diabetic neuropathy; Global = patient-reported global improvement; 40% = at least 40% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 853

854 Table 36 Characteristics of included studies: comparative trials and 855 combination therapy (randomised controlled trials)

Author Study Condition T1 T2 Titration or fixed Key outcomes period dosage (mg/day) Cross-class head-to-head comparison TCAs vs anti-epileptics Chandra et 9 weeks PHN Nortriptyline Gabapentin Nort: 50–100 50%, Mean al. (2006) Gaba: 900–2700 change in pain Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 67 of 150

relief score, AEs Achar et al. 8 weeks PHN Amitriptyline Pregabalin Ami: 25 50% (2010) Pre: 150 Within-class head-to-head comparison TCAs vs TCAs Watson et al. 5 weeks PHN Amitriptyline Nortriptyline Ami: 20 to max AEs (1998) Nort: 20 to max Combination therapy Anti-epileptics + antidepressants vs anti-epileptics vs antidepressants Achar et al. 8 weeks PHN Pregabalin Pregabalin Combination: 50% (2010) + Pre 150 + Ami 25 Amitriptyline Pre: 150 Achar et al. 8 weeks PHN Pregabalin Amitriptyline Combination: 50% (2010) + Pre 150 + Ami 25 Amitriptyline Ami: 25 T1 = treatment 1; T2 = treatment 2; PHN = post-herpetic neuralgia; 50% = at least 50% pain reduction; AEs = adverse effects. 856

857 Summary profiles 858 Meta-analyses were conducted based on the methodology stated in section 859 3.1 and presented in the following summary profiles based on individual 860 pharmacological treatments (for full GRADE profiles, see appendix XXX).

861 Antidepressants 862 Table 37 Summary profile – amitriptyline as monotherapy (placebo- 863 controlled trials) No of Relative risk Amitriptyline Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 11 16/34 4/25 RR 2.94 (1.12 to 7.73) 31 more per 100 (from 2 (47.1%) (16%) more to 100 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 6 to 8 weeks) 22 6/74 3/75 RR 1.88 (0.54 to 6.62) 4 more per 100 (from 2 (8.1%) (4%) fewer to 22 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 6 weeks) 11 11/62 15/62 RR 0.73 (0.37 to 1.47) 7 fewer per 100 (from VERY LOW (17.7%) (24.2%) 15 fewer to 11 more) Primary outcome: dry mouth (adverse effects) (follow-up 6 weeks) 11 38/62 24/62 RR 1.58 (1.09 to 2.29) 22 more per 100 (from 3 (61.3%) (38.7%) more to 50 more) VERY LOW Primary outcome: sedation (adverse effects) (follow-up 6 weeks) 11 38/62 24/62 RR 1.58 (1.09 to 2.29) 22 more per 100 (from 3 (61.3%) (38.7%) more to 50 more) VERY LOW Primary outcome: any adverse effects: unspecified (follow-up 6 weeks) 11 55/62 45/62 RR 1.22 (1.02 to 1.46) 16 more per 100 (from 1 (88.7%) (72.6%) more to 33 more) LOW 1 Max et al. (1988). 2 Graff-Radford et al. (2000); Max et al. (1988). 864

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 68 of 150

865 Table 38 Summary profile – desipramine as monotherapy (placebo- 866 controlled trials) No of Relative risk Desipramine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 11 12/19 2/19 RR 6.00 (1.55 to 53 more per 100 (from 6 (63.2%) (10.5%) 23.26) more to 100 more) MODERATE Primary outcome: no. of withdrawals due to adverse effects (follow-up 6 weeks) 11 5/19 3/19 RR 1.67 (0.46 to 6.01) 11 more per 100 (from 9 (26.3%) (15.8%) fewer to 79 more) VERY LOW Primary outcome: dizziness (adverse effects) (follow-up 6 weeks) 11 3/19 2/19 RR 1.50 (0.28 to 7.99) 5 more per 100 (from 8 (15.8%) (10.5%) fewer to 74 more) VERY LOW Primary outcome: dry mouth (adverse effects) (follow-up 6 weeks) 11 11/19 5/19 RR 2.20 (0.95 to 5.12) 32 more per 100 (from 1 (57.9%) (26.3%) fewer to 100 more) VERY LOW Primary outcome: sedation (adverse effects) (follow-up 6 weeks) 11 3/19 0/19 RR 7.00 (0.39 to – (15.8%) (0%) 126.92) VERY LOW Primary outcome: any adverse effects: unspecified (follow-up 6 weeks) 11 19/19 15/19 RR 1.26 (0.98 to 1.61) 21 more per 100 (from 2 VERY (100%) (78.9%) fewer to 48 more) LOW 1 Kishore-Kumar et al. (1990). 867

868 Anti-epileptics 869 Table 39 Summary profile – gabapentin as monotherapy (placebo- 870 controlled trials) No of Relative risk Gabapentin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 50% pain reduction (follow-up 7 weeks) 11 59/178 13/94 RR 2.40 (1.39 to 4.14) 19 more per 100 (from 5 (33.1%) (13.8%) more to 43 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 7 to 8 weeks) 22 133/299 38/207 RR 2.52 (1.29 to 4.94) 28 more per 100 (from 5 (44.5%) (18.4%) more to 72 more) LOW Primary outcome: number of withdrawals due to adverse effects (follow-up 7 to 8 weeks) 22 49/336 18/227 RR 1.87 (1.1 to 3.19) 7 more per 100 (from 1 (14.6%) (7.9%) more to 17 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 7 weeks) 11 72/223 11/111 RR 3.26 (1.8 to 5.89) 22 more per 100 (from 8 (32.3%) (9.9%) more to 48 more) VERY LOW Primary outcome: somnolence (adverse effects) (follow-up 7 weeks) 11 42/223 7/111 RR 2.99 (1.39 to 6.43) 13 more per 100 (from 2 (18.8%) (6.3%) more to 34 more) VERY LOW Primary outcome: any adverse effects: unspecified (follow-up 7 to 8 weeks) 22 192/336 63/227 RR 2.04 (1.62 to 2.58) 29 more per 100 (from (57.1%) (27.8%) 17 more to 44 more) MODERATE 1 Rice et al. (2001). 2 Rice et al. (2001); Rowbotham et al. (1998) 871 Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 69 of 150

872 Table 40 Summary profile – pregabalin as monotherapy (placebo- 873 controlled trials) No of Relative risk Pregabalin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 4 to 13 weeks) 31 251/451 59/233 RR 2.30 (1.82 to 2.91) 33 more per 100 (from (55.7%) (25.3%) 21 more to 48 more) HIGH Primary outcome: patient-reported 50% pain reduction (follow-up 4 to 13 weeks) 42 217/608 46/314 RR 2.63 (1.97 to 3.52) 24 more per 100 (from (35.7%) (14.6%) 14 more to 37 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 8 to 13 weeks) 23 107/340 21/140 RR 2.11 (1.38 to 3.22) 17 more per 100 (from 6 (31.5%) (15%) more to 33 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 13 weeks) 42 110/700 21/348 RR 2.68 (1.38 to 5.21) 10 more per 100 (from 2 (15.7%) (6%) more to 25 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 4 to 13 weeks) 42 184/700 37/348 RR 2.40 (1.49 to 3.84) 15 more per 100 (from 5 (26.3%) (10.6%) more to 30 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 4 to 13 weeks) 42 121/700 18/348 RR 3.57 (2.2 to 5.79) 13 more per 100 (from 6 (17.3%) (5.2%) more to 25 more) MODERATE Primary outcome: gait disturbances (adverse effects) (follow-up 4 to 13 weeks) 34 17/543 1/267 RR 5.31 (1.24 to 2 more per 100 (from 0 (3.1%) (0.37%) 22.74) more to 8 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 4 to 8 weeks) 25 198/268 92/174 RR 1.46 (1.25 to 1.71) 24 more per 100 (from (73.9%) (52.9%) 13 more to 38 more) MODERATE 1 Dworkin et al. (2003); Stacey et al. (2008); van Seventer et al. (2006). 2 Dworkin et al. (2003); Sabatowski et al. (2004); Stacey et al. (2008); van Seventer et al. (2006). 3 Sabatowski et al. (2004); van Seventer et al. (2006). 4 Dworkin et al. (2003); Stacey et al. (2008); van Seventer et al. (2006) . 5 Dworkin et al. (2003); Stacey et al. (2008). 874 875 Opioid analgesics 876 Table 41 Summary profile – tramadol as monotherapy (placebo- 877 controlled trials) No of Relative risk Tramadol Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 50% pain reduction 11 41/53 31/55 RR 1.37 (1.04 to 1.81) 21 more per 100 (from 2 (77.4%) (56.4%) more to 46 more) MODERATE 1 Boureau et al. (2003) 878 879 Note: no study on tramadol as monotherapy that reported the critical 880 outcomes on adverse effects was identified or met the inclusion and exclusion 881 criteria.

882

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 70 of 150

883 Topical treatments 884 Table 42 Summary profile – topical capsaicin (8% patch) as 885 monotherapy (placebo-controlled trials) No of Capsaicin 8% Relative risk Control Absolute risk Quality studies patch (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 8 to 12 weeks) 41 324/741 188/531 RR 1.26 (1.09 to 1.45) 9 more per 100 (from 3 (43.7%) (35.4%) more to 16 more) HIGH Primary outcome: patient-reported 50% pain reduction (follow-up 8 to 12 weeks) 32 158/536 70/334 RR 1.43 (1 to 2.04) 9 more per 100 (from 0 (29.5%) (21%) more to 22 more) LOW Primary outcome: patient-reported global improvement/impression of change (follow-up 8 to 12 weeks) 41 360/741 182/531 RR 1.39 (1.21 to 1.59) 13 more per 100 (from 7 (48.6%) (34.3%) more to 20 more) HIGH Primary outcome: number of withdrawals due to adverse effects (follow-up 8 to 12 weeks) 41 6/742 3/531 RR 1.32 (0.38 to 4.68) 0 more per 100 (from 0 (0.81%) (0.56%) fewer to 2 more) LOW Primary outcome: burning sensation (adverse effects) (follow-up 12 weeks) 13 3/103 0/53 RR 3.63 (0.19 to - (2.9%) (0%) 60.09) LOW Primary outcome: site pain (adverse effects) (follow-up 8 to 12 weeks) 41 253/742 104/531 RR 2.18 (1.53 to 3.11) 23 more per 100 (from (34.1%) (19.6%) 10 more to 41 more) MODERATE Primary outcome: site pruritus (adverse effects) (follow-up 8 to 12 weeks) 41 50/742 24/531 RR 1.18 (0.75 to 1.87) 1 more per 100 (from 1 (6.7%) (4.5%) fewer to 4 more) LOW Primary outcome: site papules (adverse effects) (follow-up 8 to 12 weeks) 34 38/639 13/478 RR 2.53 (1.39 to 4.61) 4 more per 100 (from 1 (5.9%) (2.7%) more to 10 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 to 12 weeks) 41 618/742 422/531 RR 1.14 (1.08 to 1.2) 11 more per 100 (from 6 (83.3%) (79.5%) more to 16 more) HIGH 1 Backonja et al. (2008); Irving et al. (2011); Webster et al. (2010a); Webster et al. (2010b). 2 Irving et al. (2011); Webster et al. (2010a); Webster et al. (2010b). 3 Webster et al. (2010a). 4 Backonja et al. (2008); Irving et al. (2011); Webster et al. (2010b). 886

887 Table 43 Summary profile – topical capsaicin (0.075% cream) as 888 monotherapy (placebo-controlled trials) No of Capsaicin Relative risk Control Absolute risk Quality studies 0.075% cream (95% CI) Primary outcome: patient-reported 40% pain reduction (follow-up 6 weeks) 11 7/16 1/16 RR 7.00 (0.97 to 38 more per 100 (from 0 (43.8%) (6.3%) 50.57) fewer to 100 more) MODERATE Primary outcome: burning (adverse effects) (follow-up mean 6 weeks) 22 50/90 25/85 RR 1.88 (1.30 to 2.72) 26 more per 100 (from 9 (55.6%) (29.4%) more to 51 more) LOW Primary outcome: number of withdrawals due to adverse effects (follow-up 6 weeks) 13 13/74 1/69 RR 12.12 (1.63 to 16 more per 100 (from 1 (17.6%) (1.4%) 90.23) more to 100 more) VERY LOW 1 Bernstein et al. (1989). 2 Bernstein et al. (1989); Watson et al. (1993). 3 Watson et al. (1993). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 71 of 150

889

890 Table 44 Summary profile – topical lidocaine (5% patch) as monotherapy 891 (placebo-controlled trials) Topical No of Relative risk lidocaine 5% Control Absolute risk Quality studies (95% CI) patch Other reported pain outcome: pain relief (scale: Neuropathic Pain Scale) (follow-up 3 weeks) 11 67 29 Treatment = −15.3 (17.9) Control = −7.7 (14.2) LOW p = 0.043 1 Galer et al. (2002). 892

893 Note: no study on topical lidocaine (5% patch) as monotherapy that reported 894 the critical outcomes on pain and adverse effects was identified or met the 895 inclusion and exclusion criteria.

896 Head-to-head comparative trials (monotherapy) 897 Table 45 Summary profile – amitriptyline vs pregabalin as monotherapy 898 (comparative trials) No of Relative risk Amitriptyline Pregabalin Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 75% pain reduction (follow-up 8 weeks) 11 2/15 8/15 RR 0.25 (0.06 to 0.99) 40 fewer per 100 (from (13.3%) (53.3%) 1 fewer to 50 fewer) LOW 1 Achar et al. (2010). 899

900 Note: no study on amitriptyline vs pregabalin as monotherapy that reported 901 the critical outcomes on adverse effects was identified or met the inclusion 902 and exclusion criteria.

903 Table 46 Summary profile – amitriptyline vs nortriptyline as 904 monotherapy (comparative trials) No of Relative risk Amitriptyline Nortriptyline Absolute risk Quality studies (95% CI) Primary outcome: dry mouth (adverse effects) (follow-up 5 weeks) 11 28/33 26/33 RR 1.08 (0.86 to 1.35) 6 more per 100 (from 11 (84.8%) (78.8%) fewer to 28 more) VERY LOW Primary outcome: dizziness (adverse effects) (follow-up 5 weeks) 11 3/33 1/33 RR 3.00 (0.33 to 27.4) 6 more per 100 (from 2 (9.1%) (3%) fewer to 80 more) VERY LOW Primary outcome: drowsiness (adverse effects) (follow-up 5 weeks) 11 4/33 6/33 RR 0.67 (0.21 to 2.13) 6 fewer per 100 (from (12.1%) (19.4%) 15 fewer to 22 more) VERY LOW Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 72 of 150

Primary outcome: any adverse effects: unspecified (follow-up 5 weeks) 11 31/33 31/33 RR 1.00 (0.88 to 1.13) 0 fewer per 100 (from (93.9%) (93.9%) 11 fewer to 12 more) VERY LOW 1 Watson et al. (1998). 905

906 Note: no study on amitriptyline vs nortriptyline as monotherapy that reported 907 the critical outcomes on pain was identified or met the inclusion and exclusion 908 criteria

909 Table 47 Summary profile – nortriptyline vs gabapentin as monotherapy 910 (comparative trials) No of Relative risk Nortriptyline Gabapentin Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 50% pain reduction (follow-up 9 weeks) 11 9/36 7/34 RR 1.21 (0.51 to 2.90) 4 more per 100 (from 10 (25%) (20.6%) fewer to 39 more) MODERATE Primary outcome: somnolence (adverse effects) (follow-up 9 weeks) 11 6/36 4/34 RR 1.42 (0.44 to 4.59) 5 more per 100 (from 7 (16.7%) (11.8%) fewer to 42 more) VERY LOW Primary outcome: dry mouth (adverse effects) (follow-up 9 weeks) 11 18/36 0/34 RR ∞ (∞) – (50%) (0%) VERY LOW 1 Chandra et al. (2006). 911

912 Head-to-head comparative trials (combination therapy) 913 Table 48 Summary profile – pregabalin + amitriptyline as combination 914 therapy vs amitriptyline alone (comparative trials) No of Pregabalin + Relative risk Amitriptyline Absolute risk Quality studies amitriptyline (95% CI) Primary outcome: patient-reported 75% pain reduction (follow-up 8 weeks) 11 11/15 2/15 RR 5.5 (1.46 to 20.71) 60 more per 100 (from 6 (73.3%) (13.3%) more to 100 more) MODERATE 1 Achar et al. (2010). 915

916 Note: no study on pregabalin + amitriptyline as combination therapy vs 917 amitriptyline alone that reported the critical outcomes on adverse effects was 918 identified or met the inclusion and exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 73 of 150

919 Table 49 Summary profile – pregabalin + amitriptyline as combination 920 therapy vs pregabalin alone (comparative trials) No of Pregabalin + Relative Pregabalin Absolute Quality studies amitriptyline (95% CI) Primary outcome: patient-reported 75% pain reduction (follow-up 8 weeks) 11 11/15 8/15 RR 1.38 (0.78 to 2.41) 20 more per 100 (from (73.3%) (53.3%) 12 fewer to 75 more) MODERATE 1 Achar et al. (2010). 921

922 Note: No study on pregabalin + amitriptyline as combination therapy vs 923 pregabalin alone that reported the critical outcomes on adverse effects was 924 identified or met the inclusion and exclusion criteria.

925 3.3.3 Evidence statements

926 For details of how the evidence is graded, see ‘The guidelines manual’.

927 3.3.3.1 No study on clomipramine, dosulepin (dothiepin), doxepin, 928 imipramine, lofepramine, trimipramine, citalopram, fluoxetine, 929 paroxetine, sertraline, duloxetine, venlafaxine, carbamazepine, 930 lamotrigine, oxcarbazepine, phenytoin, sodium valproate, 931 topiramate, buprenorphine, co-codamol, codeine phosphate, co- 932 dydramol, dihydrocodeine, fentanyl, morphine and oxycodone was 933 identified or met the inclusion and exclusion criteria for PHN.

934 Antidepressants as monotherapy against placebo 935 Amitriptyline (linked to table 37) 936 Critical outcomes (pain) 937 3.3.3.2 Moderate quality evidence from one study with 59 patients with 938 PHN, showed that amitriptyline is more effective than placebo in 939 achieving global improvement/impression of change from baseline 940 up to 6 weeks’ follow-up.

941 Critical outcomes (adverse effects) 942 3.3.3.3 Low quality evidence from two studies with 149 patients with PHN, 943 showed that there is no significant difference between patients on 944 amitriptyline and placebo withdrawing from studies due to adverse 945 effects from baseline up to 8 weeks’ follow-up. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 74 of 150

946 3.3.3.4 Very low quality evidence from one study with 124 patients with 947 PHN, showed that there is no significant difference between 948 patients on amitriptyline and placebo in experiencing dizziness from 949 baseline up to 8 weeks’ follow-up.

950 3.3.3.5 Very low quality evidence from one study with 124 patients with 951 PHN, showed that patients on amitriptyline are more likely to 952 experience dry mouth and sedation compared with placebo from 953 baseline up to 6 weeks’ follow-up.

954 3.3.3.6 Low quality evidence from one study with 124 patients with PHN, 955 showed that patients on amitriptyline are more likely to experience 956 any adverse effects (unspecified) compared with placebo from 957 baseline up to 6 weeks’ follow-up.

958 Desipramine (linked to table 38) 959 Critical outcomes (pain) 960 3.3.3.7 Moderate quality evidence from one study with 38 patients with 961 PHN, showed that desipramine is more effective than placebo in 962 achieving global improvement/impression of change from baseline 963 up to 6 weeks’ follow-up.

964 Critical outcomes (adverse effects) 965 3.3.3.8 Very low quality evidence from one study with 38 patients with 966 PHN, showed that there is no significant difference between 967 patients on desipramine and placebo withdrawing from studies due 968 to adverse effects, or experiencing dizziness, dry mouth, sedation 969 and any adverse effects (unspecified) from baseline up to 6 weeks’ 970 follow-up.

971 Anti-epileptics as monotherapy against placebo 972 Gabapentin (linked to table 39) 973 Critical outcomes (pain) 974 3.3.3.9 Moderate quality evidence from one study with 272 patients with 975 PHN, showed that gabapentin is more effective than placebo in

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 75 of 150

976 achieving at least 50% pain reduction from baseline up to 7 weeks’ 977 follow-up.

978 3.3.3.10 Low quality evidence from two studies with 506 patients with PHN, 979 showed that gabapentin is more effective than placebo in achieving 980 global improvement/impression of change from baseline up to 981 8 weeks’ follow-up.

982 Critical outcomes (adverse effects) 983 3.3.3.11 Low quality evidence from two studies with 563 patients with PHN, 984 showed that patients on gabapentin are more likely to withdraw 985 from studies due to adverse effects compared with placebo from 986 baseline up to 8 weeks’ follow-up.

987 3.3.3.12 Moderate quality evidence from two studies with 563 patients with 988 PHN, showed that patients on gabapentin are more likely to 989 experience any adverse effects (unspecified) compared with 990 placebo from baseline up to 8 weeks’ follow-up.

991 3.3.3.13 Very low quality evidence from one study with 334 patients with 992 PHN, showed that patients on gabapentin are more likely to 993 experience dizziness and somnolence compared with placebo from 994 baseline up to 7 weeks’ follow-up.

995 Pregabalin (linked to table 40) 996 Critical outcomes (pain) 997 3.3.3.14 High quality evidence from three studies with 684 patients with 998 PHN, showed that pregabalin is more effective than placebo in 999 achieving at least 30% pain reduction from baseline up to 1000 13 weeks’ follow-up.

1001 3.3.3.15 Moderate quality evidence from four studies with 922 patients with 1002 PHN, showed that pregabalin is more effective than placebo in 1003 achieving at least 50% pain reduction from baseline up to 1004 13 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 76 of 150

1005 3.3.3.16 Moderate quality evidence from two studies with 480 patients with 1006 PHN, showed that pregabalin is more effective than placebo in 1007 achieving at global improvement/impression of change from 1008 baseline up to 13 weeks’ follow-up.

1009 Critical outcomes (adverse effects) 1010 3.3.3.17 Low quality evidence from four studies with 1048 patients with 1011 PHN, showed that patients on pregabalin are more likely to 1012 withdraw from studies due to adverse effects, and experience 1013 dizziness compared with placebo from baseline up to 13 weeks’ 1014 follow-up.

1015 3.3.3.18 Moderate quality evidence from four studies with 1048 patients with 1016 PHN, showed that patients on pregabalin are more likely to 1017 experience somnolence compared with placebo from baseline up to 1018 13 weeks’ follow-up.

1019 3.3.3.19 Low quality evidence from three studies with 810 patients with 1020 PHN, showed that patients on pregabalin are more likely to 1021 experience gait disturbances compared with placebo from baseline 1022 up to 13 weeks’ follow-up.

1023 3.3.3.20 Moderate quality evidence from two studies with 442 patients with 1024 PHN, showed that patients on pregabalin are more likely to 1025 experience any adverse effects (unspecified) compared with 1026 placebo from baseline up to 8 weeks’ follow-up.

1027 Opioid analgesics as monotherapy against placebo 1028 Tramadol (linked to table 41) 1029 Critical outcomes (pain) 1030 3.3.3.21 Moderate quality evidence from one study with 108 patients with 1031 PHN, showed that tramadol is more effective than placebo in 1032 achieving at least 50% pain reduction from baseline up to 1033 13 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 77 of 150

1034 Critical outcomes (adverse effects) 1035 3.3.3.22 No study on tramadol as monotherapy that reported the critical 1036 outcomes on adverse effects was identified or met the inclusion 1037 and exclusion criteria.

1038 Topical treatments as monotherapy against placebo 1039 Topical capsaicin (8% patch) (linked to table 42) 1040 Critical outcomes (pain) 1041 3.3.3.23 High quality evidence from four studies with 1272 patients with 1042 PHN, showed that topical capsaicin (8% patch) is more effective 1043 than placebo in achieving at least 30% pain reduction and global 1044 improvement/impression of change from baseline up to 12 weeks’ 1045 follow-up.

1046 3.3.3.24 Low quality evidence from three studies with 870 patients with 1047 PHN, showed that topical capsaicin (8% patch) is more effective 1048 than placebo in achieving at least 50% pain reduction from baseline 1049 up to 12 weeks’ follow-up.

1050 Critical outcomes (adverse effects) 1051 3.3.3.25 Low quality evidence from four studies with 1273 patients with 1052 PHN, showed that there is no significant difference between 1053 patients on topical capsaicin (8% patch) and placebo withdrawing 1054 from studies due to adverse effects and experiencing site pruritus 1055 from baseline up to 12 weeks’ follow-up.

1056 3.3.3.26 Low quality evidence from one study with 156 patients with PHN, 1057 showed that there is no significant difference between patients on 1058 topical capsaicin (8% patch) and placebo in experiencing burning 1059 sensation from baseline up to 12 weeks’ follow-up.

1060 3.3.3.27 Moderate quality evidence from four studies with 1273 patients with 1061 PHN, showed that patients on topical capsaicin (8% patch) are 1062 more likely to experience site pain compared with placebo from 1063 baseline up to 12 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 78 of 150

1064 3.3.3.28 High quality evidence from four studies with 1273 patients with 1065 PHN, showed that patients on topical capsaicin (8% patch) are 1066 more likely to experience any adverse effects (unspecified) 1067 compared with placebo from baseline up to 12 weeks’ follow-up.

1068 3.3.3.29 Low quality evidence from three studies with 1117 patients with 1069 PHN, showed that patients on topical capsaicin (8% patch) are 1070 more likely to experience site papules compared with placebo from 1071 baseline up to 12 weeks’ follow-up.

1072 Topical capsaicin (0.075% cream) (linked to table 43) 1073 Critical outcomes (pain) 1074 3.3.3.30 Moderate quality evidence from one study with 32 patients with 1075 PHN, showed that there is no significant difference between 1076 patients on topical capsaicin (0.075% cream) and placebo in 1077 achieving 40% pain reduction from baseline up to 6 weeks’ follow- 1078 up.

1079 Critical outcomes (adverse effects) 1080 3.3.3.31 Low quality evidence from two studies with 175 patients with PHN, 1081 showed that patients on topical capsaicin (0.075% cream) are more 1082 likely to experience burning compared with placebo from baseline 1083 up to 6 weeks’ follow-up.

1084 3.3.3.32 Very low quality evidence from one study with 143 patients with 1085 PHN, showed that patients on topical capsaicin (0.075% cream) 1086 are more likely to withdraw from studies due to adverse effects 1087 compared with placebo from baseline up to 6 weeks’ follow-up.

1088 Topical lidocaine (5% patch) (linked to table 44) 1089 Critical outcomes (pain) 1090 3.3.3.33 No study on topical lidocaine (5% patch) as monotherapy that 1091 reported the critical outcomes on pain was identified or met the 1092 inclusion and exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 79 of 150

1093 Critical outcomes (adverse effects) 1094 3.3.3.34 No study on topical lidocaine (5% patch) as monotherapy that 1095 reported the critical outcomes on adverse effects was identified or 1096 met the inclusion and exclusion criteria.

1097 Other reported pain outcomes 1098 3.3.3.35 Low quality evidence from one study with 96 patients with PHN, 1099 showed that patients on topical lidocaine (5% patch) are more likely 1100 to have better scores in pain relief scale compared with placebo 1101 from baseline up to 3 weeks’ follow-up.

1102 Head-to-head comparative trials (monotherapy) 1103 Amitriptyline vs pregabalin (linked to table 45) 1104 Critical outcomes (pain) 1105 3.3.3.36 Low quality evidence from one study with 30 patients with PHN, 1106 showed that there is no significant difference between patients on 1107 amitriptyline and patients on pregabalin in achieving 75% pain 1108 reduction from baseline up to 8 weeks’ follow-up.

1109 Critical outcomes (adverse effects) 1110 3.3.3.37 No study on amitriptyline vs pregabalin as monotherapy that 1111 reported the critical outcomes on adverse effects was identified or 1112 met the inclusion and exclusion criteria.

1113 Amitriptyline vs nortriptyline (linked to table 46) 1114 Critical outcomes (pain) 1115 3.3.3.38 No study on amitriptyline vs nortriptyline as monotherapy that 1116 reported the critical outcomes on pain was identified or met the 1117 inclusion and exclusion criteria.

1118 Critical outcomes (adverse effects) 1119 3.3.3.39 Very low quality evidence from one study with 66 patients with 1120 PHN, showed that there is no significant difference between 1121 patients on amitriptyline and patients on nortriptyline in 1122 experiencing dry mouth, dizziness, drowsiness and any adverse 1123 effects (unspecified) from baseline up to 5 weeks’ follow-up. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 80 of 150

1124 Nortriptyline vs gabapentin (linked to table 47) 1125 Critical outcomes (pain) 1126 3.3.3.40 Moderate quality evidence from one study with 70 patients with 1127 PHN, showed that there is no significant difference between 1128 patients on nortriptyline and patients on gabapentin in achieving at 1129 least 50% pain reduction from baseline up to 9 weeks’ follow-up.

1130 Critical outcomes (adverse effects) 1131 3.3.3.41 Very low quality evidence from one study with 70 patients with 1132 PHN, showed that there is no significant difference between 1133 patients on nortriptyline and patients on gabapentin in experiencing 1134 somnolence and dry mouth from baseline up to 9 weeks’ follow-up.

1135 Head-to-head comparative trials (combination therapy) 1136 Pregabalin + amitriptyline as combination therapy vs amitriptyline alone 1137 (linked to table 48) 1138 Critical outcomes (pain) 1139 3.3.3.42 Moderate quality evidence from one study with 30 patients with 1140 PHN, showed that pregabalin + amitriptyline is more effective than 1141 amitriptyline alone in achieving 75% pain reduction from baseline 1142 up to 8 weeks’ follow-up.

1143 Critical outcomes (adverse effects) 1144 3.3.3.43 No study on pregabalin + amitriptyline as combination therapy vs 1145 amitriptyline alone that reported the critical outcomes on adverse 1146 effects was identified or met the inclusion and exclusion criteria.

1147 Pregabalin + amitriptyline as combination therapy vs pregabalin alone 1148 (linked to table 49) 1149 Critical outcomes (pain) 1150 3.3.3.44 Moderate quality evidence from one study with 30 patients with 1151 PHN, showed that there is no significant difference between 1152 pregabalin + amitriptyline and pregabalin alone in achieving 75% 1153 pain reduction from baseline up to 8 weeks follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 81 of 150

1154 Critical outcomes (adverse effects) 1155 3.3.3.45 No study on pregabalin + amitriptyline as combination therapy vs 1156 pregabalin alone that reported the critical outcomes on adverse 1157 effects was identified or met the inclusion and exclusion criteria.

1158 3.3.4 Health economic modelling

1159 The analysis presented results in terms of decreasing mean net monetary 1160 benefit associated with each drug at a threshold of £20,000 and £30,000 per 1161 QALY gained with all comparisons to placebo.

1162 The cost effectiveness results for the PHN model are presented in tables 50 1163 and 51. These results indicate that, of the single doses considered, 1164 gabapentin 3600 mg is the most cost-effective option at both thresholds; 1165 pregabalin 150 mg is the second most cost effective option, followed by 1166 pregabalin 300 mg. The flexible dosing analysis indicates that gabapentin is 1167 the most cost effective. Pregabalin and oxycodone are associated with very 1168 low probabilities of being cost effective.

1169 Table 50 PHN incremental cost effectiveness results Mean incremental net benefit (£) per Drug person at a threshold per QALY of: £20,000 £30,000 Single dose comparators Gabapentin 3600 mg 2010 2974 Pregabalin 150 mg 1268 2260 Pregabalin 300 mg 846 1616 Gabapentin 1800 mg 976 1430 Gabapentin 2400 mg 964 1430 Oxycodone 60 mg 400 914 Pregabalin 600 mg 194 616 Lidocaine 5% patch −1248 −1015 Single and flexible-dose comparators Gabapentin flexible-dose (1800–3600 mg) 1303 1921 Pregabalin flexible dose (150–600 mg) 707 1403 Oxycodone 60 mg 400 914 Lidocaine 5% patch −1248 −1015 1170 1171

1172 Table 51 PHN probabilistic results

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 82 of 150

Single dose analysis Flexible dose analysis Probability of being the most Probability of being the most cost-effective drug at a cost-effective drug at a Drug threshold per QALY of: Drug threshold per QALY of: £20,000 £30,000 £20,000 £30,000 Gabapentin Gabapentin 3600 mg flexible dose 0.656 0.573 0.8885 0.719 (1800– 3600 mg) Pregabalin Pregabalin 150 mg 0.172 0.252 flexible dose 0.0605 0.1815 (150–600 mg) Pregabalin Oxycodone 0.060 0.092 0.051 0.0995 300 mg 60 mg Gabapentin Lidocaine 5% 0.059 0.041 0 0 1800 mg patch Gabapentin Placebo 0.046 0.031 0 0 2400 mg Oxycodone 0.005 0.008 60 mg Pregabalin 0.003 0.005 600 mg Lidocaine 5% 0 0 patch Placebo 0 0 1173

1174 Sensitivity and scenario analyses 1175 Numerous sensitivity analyses were conducted to explore how the model’s 1176 inputs affected its results and, in particular, the extent to which single 1177 parameters would need to be altered before different options became cost 1178 effective (‘threshold analysis’). The key variables that affected the decision are 1179 summarised below.

1180 Price changes – pregabalin 1181 When the price of pregabalin was reduced to 20% of its current price (or a 1182 reduction in 80%) for the single-dose analysis pregabalin became the 1183 intervention with the highest mean incremental net benefit at £20,000 and 1184 £30,000 thresholds. In the flexible-dose analysis, at a £20,000 threshold, 1185 pregabalin again had the highest mean incremental net benefit. However, at a 1186 £30,000 threshold the price only had to fall to 40% of its current price for it to 1187 become the most cost-effective option.

1188 Reduced pill burden – gabapentin 1189 If the number of pills for gabapentin were reduced, so that the overall dose 1190 remained the same, but pills of larger dose sizes were used, and all other

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 83 of 150

1191 drugs were kept the same, this resulted in the highest drug costs to 1192 gabapentin. The outcome of this change was that pregabalin 150 mg had the, 1193 marginally (£50), highest net benefit at £30,000 threshold and a very similar 1194 probability of being cost effective (0.03 difference in probabilities). These 1195 results were repeated at £20,000 threshold but the differences were reduced. 1196 These results were not replicated for the flexible-dose analysis. Gabapentin 1197 remained the most cost-effective option, and at the £20,000 threshold the 1198 probabilities were very different, although at the £30,000 threshold the 1199 differences were much smaller.

1200 Lidocaine patches 1201 Even if the price of lidocaine patches and the number of patches reduced to 1202 1.03 patches, it remained the least cost-effective option.

1203 Oxycodone 1204 The price of oxycodone needs to be reduced by 40% (60%) of current prices 1205 at £30,000 (£20,000) thresholds for oxycodone to become as cost effective as 1206 pregabalin.

1207 Time horizon 1208 If the time horizon was reduced to 1 year, then the net benefits were reduced 1209 across all the treatments, but the difference between gabapentin and 1210 pregabalin increased.

1211 Utilities 1212 Changing the assumptions around utilities resulted in the differences between 1213 gabapentin and pregabalin being reduced, particularly for the single-dose 1214 analysis.

1215 Health economics evidence statements – PHN 1216 3.3.4.1 Partially applicable evidence from one study with minor limitations, 1217 showed that gabapentin was the most cost effective treatment for 1218 PHN compared with oxycodone, pregabalin and lidocaine 5% 1219 patch.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 84 of 150

1220 3.3.5 Evidence to recommendations

1221 Because the evidence-base for post-herpetic neuralgia (PHN) is similar to that 1222 for other neuropathic pain conditions (that all other neuropathic pain 1223 conditions apart from PDN and PHN), the GDG felt that it may be more 1224 appropriate to discuss the evidence for PHN alongside the evidence for other 1225 neuropathic pain conditions. Hence, for evidence to recommendations for 1226 PHN, please see section 3.4.5.

1227

1228 3.3.6 Recommendations and research recommendations

1229 Recommendations Please see section 3.4.6.

1230

1231 Research recommendations 1232 See appendix B for full details of research recommendations.

1233 3.4 Other neuropathic pain conditions

1234 3.4.1 Review questions

1235 Review question 1 1236 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 1237 analgesics, topical lidocaine and topical capsaicin as monotherapy (against 1238 placebo) for the management of neuropathic pain condition (other than PDN 1239 and PHN) in adults in non-specialist settings?

1240 Review question 2 1241 What is the clinical efficacy of antidepressants, anti-epileptics, opioid 1242 analgesics, topical lidocaine and topical capsaicin as combination therapy 1243 (against monotherapy or other combination therapy) for the management of 1244 neuropathic pain (other than PDN and PHN) in adults in non-specialist 1245 settings?

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 85 of 150

1246 Review question 3 1247 What is the clinical efficacy of antidepressants, compared with anti-epileptics, 1248 opioid analgesics, topical lidocaine and topical capsaicin (and vice-versa) as 1249 monotherapy for the management of neuropathic pain (other than PDN and 1250 PHN) in adults in non-specialist settings?

1251 3.4.2 Evidence review

1252 A total of 60 randomised controlled trials were included for other neuropathic 1253 pain conditions (other than PDN and PHN). Of the 34 listed included 1254 pharmacological treatments in (Table 4), no study was identified or met the 1255 inclusion and exclusion criteria for the following pharmacological treatments 1256 (see table 52).

1257 For the characteristics of included studies please see Table 53 to 57.

1258 Table 52 Pharmacological treatments for which no studies met the 1259 inclusion and exclusion criteria for other neuropathic pain conditions Drug class: subclass Drug Antidepressants: tricyclic antidepressants (TCAs) Clomipramine Desipramine Dosulepin (dothiepin) Doxepin Lofepramine Trimipramine Antidepressants: selective serotonin reuptake Citalopram inhibitors (SSRIs) Fluoxetine Paroxetine Sertraline Anti-epileptics (anticonvulsants) Oxcarbazepine Phenytoin Sodium valproate Opioid analgesics Buprenorphine Co-codamol Codeine phosphate Co-dydramol Dihydrocodeine Fentanyl Topical treatments Topical capsaicin Topical lidocaine 1260 Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 86 of 150

1261 Table 53 Characteristics of included studies: antidepressants (placebo- 1262 controlled trials)

Author Study Condition Treatment Titration or Mean Outcomes period (oral) fixed dosage dose (mg/day) (mg/day) Cardenas et 6 weeks SCI Amitriptyline 10–125 **50 Mean pain al. (2002) intensity score, AEs Rintala et al. 8 weeks SCI Amitriptyline 150 150 30%, AEs (2007) Kalso et al. 4 weeks NP cancer Amitriptyline 5–100 93.3 AEs (1996) Kautio et al. 8 weeks NP cancer Amitriptyline 10–50 46.2 Global, AEs (2008) Kieburtz et al. 9 weeks HIV-RN Amitriptyline 25–100 NR Global, AEs (1998) Leijon and 4 weeks PSP Amitriptyline 25–75 75 Global Boivie (1989) Robinson et al. 6 weeks PhanLP Amitriptyline 10–125 NR Mean pain relief (2004) score, AEs Vrethem et al. 4 weeks Poly Amitriptyline 25–75 NR 30%, Global, (1997) AEs Khoromi et al. 7 weeks Radi Nortriptyline 25–100 84 Global (2007) Sindrup et al. 4 weeks Poly Imipramine 50–150 NR Global, AEs (2003) Vranken et al. 8 weeks Central Duloxetine 60-120 99.1 Global, AEs (2011) pain Sindrup et al. 4 weeks Poly Venlafaxine 75–225 NR Global, AEs (2003) Tasmuth et al. 4 weeks NP cancer Venlafaxine 18.75–75 n/a AEs (2002) Yucel et al. 8 weeks Mixed NP Venlafaxine 75, 150 N/A Global, AEs (2005) ** = median; SCI = spinal cord injury; NP cancer = neuropathic cancer pain; HIV-RN = HIV-related neuropathy; PSP = post-stroke pain; PhanLP = phantom limb pain; Poly = polyneuropathy; Radi = radiculopathy; Mixed NP = mixed neuropathic pain; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects; NR = not reported; N/A = not applicable. 1263

1264 Table 54 Characteristics of included studies: anti-epileptics (placebo- 1265 controlled trials) Author Study Condition Treatment (oral) Titration or Outcomes period fixed dosage (mg/day) Leijon and 4 weeks PSP Carbamazepine 200–800 Global Boivie (1989) Nicol (1969) 46 months TN Carbamazepine 100–2400 Global Khoromi et al. 6 weeks Radi Topiramate 50–400 Global, AEs (2005) Simpson et al. 14 weeks HIV-RN Lamotrigine 50–300 AEs (2000) Simpson et al. 12 weeks HIV-RN Lamotrigine 25–400 Global, AEs (2003) Breuer et al. 11 weeks MS-NP Lamotrigine 25–400 30%, AEs (2007) Finnerup et al. 9 weeks SCI Lamotrigine 25–400 AEs (2002) Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 87 of 150

McCleane 8 weeks Mixed NP Lamotrigine 25–200 AEs (1999) Rao et al. 10 weeks NP cancer Lamotrigine 25–300 AEs (2008) Vestergaard et 8 weeks PSP Lamotrigine 200 AEs al. (2001) 6 weeks PhanLP Gabapentin 300–2400 Mean change in Bone et al. pain intensity (2002) score, AEs Nikolajsen et 30 days PhanLP Gabapentin 300–2400 AEs al. (2006) Smith et al. 6 weeks PhanLP Gabapentin 300–3600 Global (2005) Levendoglu et 8 weeks SCI Gabapentin 900–3600 Mean pain relief al. (2004) score, AEs Rintala et al. 8 weeks SCI Gabapentin to 3600 30%, AEs (2007) Gordh et al. 5 weeks NP-NI Gabapentin 300–2400 Global, AEs (2008) Hahn et al. 4 weeks HIV-RN Gabapentin 400–2400 AEs (2004) Rao et al. 6 weeks NP cancer Gabapentin 300–2700 AEs (2007) 8 weeks Mixed NP Gabapentin 900–2400 50%, Global, Serpell (2002) AEs Freynhagen et 12 weeks PDN, PHN Pregabalin 150–600, 300– 30%, 50%, al. (2005) 600 Global, AEs Siddall et al. 12 weeks SCI Pregabalin 150–600 30%, 50%, AEs (2006) Vranken et 4 weeks CenP Pregabalin 150–600 AEs al.(2008) Simpson et al. 14 weeks HIV-RN Pregabalin 150-600 30%, 50%, (2010) Global, AEs van Seventer 8 weeks Post-Trauma Pregabalin 150-600 30%, Global, et al. (2010) AEs Moon et al. 8 weeks Mixed NP Pregabalin 150-600 30%, 50%, (2010) Global, AEs MS-NP = multiple sclerosis neuropathic pain (central pain); NP-NI = nerve injury neuropathic pain; CenP = central pain; SCI = spinal cord injury; NP cancer = neuropathic cancer pain; HIV-RN = HIV-related neuropathy; PSP = post-stroke pain; PhanLP = phantom limb pain; Radi = radiculopathy; Mixed NP = mixed neuropathic pain; Post-Trauma = post-traumatic pain (including post-surgical pain); Global = patient- reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. TN = Trigeminal neuralgia 1266

1267 Table 55 Characteristics of included studies: opioid analgesics (placebo- 1268 controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (oral) dosage (mg/day) Arbaiza and 6 weeks NP cancer Tramadol **68.75 Mean pain intensity Vidal (2007) score, AEs Sindrup et al. 4 weeks Poly Tramadol 100–400 Mean pain intensity (1999) score, AEs Norrbrink et al. 4 weeks SCI Tramadol 150-400 Global, AEs (2009) Huse et al. 4 weeks PhanLP Morphine 70–300 50% (2001) Khoromi et al. 6 weeks Radi Morphine 15–180 Global, AEs (2007a) Wu et al. 7 weeks PhanLP Morphine 15–90 30%, 50%, AEs Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 88 of 150

(2008) **mean mg/6 hours; NP cancer = neuropathic cancer pain; PhanLP = phantom limb pain; Poly = polyneuropathy; Radi = radiculopathy; SCI = spinal cord injury ; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 1269

1270 Table 56 Characteristics of included studies: topical capsaicin and 1271 topical lidocaine (placebo-controlled trials)

Author Study Condition Treatment Titration or fixed Outcomes period (topical) dosage (times/day) Donofrio et al. 8 weeks PDN or Radi Capsaicin 0.075% cream, 4 Mean pain relief (1991) score, mean change in pain intensity score, AEs Low et al. 8 weeks Poly Capsaicin 0.075% cream, 4 Mean pain intensity (1995) score, AEs McCleane 4 weeks Mixed NP Capsaicin 0.025% cream, 3 Mean change in pain (2000) intensity score Paice et al. 4 weeks HIV-RN Capsaicin 0.075% cream, 4 AEs (2000) Watson and 6 weeks NP cancer Capsaicin 0.075% cream, 4 50%, AEs Evans (1992) Meier et al. 1 week Peri NP Lidocaine 5% patch, up to 4 Mean change in pain (2003) patches for intensity score, AEs 12 hours/day Ho et al. 1 week Mixed NP Lidocaine 5% cream, 2 Mean change in pain (2008) intensity score, AEs Cheville et al. 4 weeks PS-NP Lidocaine 5% patch, up to 3 Mean pain intensity (2009) patches for score 18 hours/day Estanislao et 2 weeks HIV-RN Lidocaine 5% gel, 1 Mean pain relief al. (2004) score NP cancer = neuropathic cancer pain; Poly = polyneuropathy; Radi = radiculopathy; HIV-RN = HIV-related neuropathy; PS-NP = postsurgical neuropathic pain; Peri NP = peripheral neuropathic pain; Mixed NP = mixed neuropathic pain; Global = patient-reported global improvement; 50% = at least 50% pain reduction; AEs = adverse effects. 1272

1273 Table 57 Characteristics of included studies: comparative trials and 1274 combination therapy (randomised controlled trials)

Author Study Condition T1 T2 Titration or Outcomes period fixed dosage (mg/day) Cross-class head-to-head comparison TCAs vs anti-epileptics Rintala et 8 SCI Amitriptyline Gabapentin Ami: max 150 30%, AEs al. (2007) weeks Gaba: max 3600 Leijon and 4 PSP Amitriptyline Carbamazepine Ami: 25–75 Global, AEs Boivie weeks Carba: 200–800 (1989) Anti-epileptics vs opioids Gatti et al. 3 Mixed NP Pregabalin Oxycodone Pre: 85.6 to max Mean pain (2009) months Oxy: 24.1 to max intensity score, AEs Anti-epileptics vs topical lidocaine Baron et 4 PDN Pregabalin Topical lidocaine Pre: 150-600 30%, 50%, Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 89 of 150

al. (2009) weeks PHN 5% Lido: 3–4 Global, AEs patches up to 12 hours/day Within-class head-to-head comparison TCAs vs SNRIs Sindrup et 4 Poly Imipramine Venlafaxine Imi: 50–150 Global, AEs al. (2003) weeks Ven: 75–225 Combination therapy Anti-epileptics + opioids vs anti-epileptics Gatti et al. 3 Mixed NP Pregabalin + Pregabalin Combination: Mean pain (2009) months oxycodone Pre 108.1 + Oxy intensity 19.4 score, AEs Pre: 85.6 to max Anti-epileptics + opioids vs opioids Gatti et al. 3 Mixed NP Pregabalin + Oxycodone Combination: Mean pain (2009) months oxycodone Pre 108.1 + Oxy intensity 19.4 score, AEs Oxy: 24.1 to max Anti-epileptics + antidepressants vs anti-epileptics vs antidepressants Gilron et 6 PDN, PHN Gabapentin Gabapentin Combination: Mean al. (2009) weeks + Gaba 3600 + change in nortriptyline Nort 100 daily pain Gaba: 3600 score Gilron et 6 PDN, PHN Gabapentin Nortriptyline Combination: Mean al. (2009) weeks + Gaba 3600 + change in nortriptyline Nort 100 daily pain Nort: 100 score T1 = treatment 1; T2 = treatment 2; Mixed NP = mixed neuropathic pain; PSP = post-stroke pain; Poly = polyneuropathy; SCI = spinal cord injury; Ami = amitriptyline; Gaba = gabapentin; Nort = nortriptyline; Carba = carbamazepine; Pre = pregabalin; Oxy = oxycodone; Cap = topical capsaicin; Lido = topical lidocaine; Imi = imipramine; Ven = venlafaxine; Global = patient-reported global improvement; 30% = at least 30% pain reduction; 50% = at least 50% pain reduction; AEs = adverse effects. 1275

1276 Summary profiles 1277 Meta-analyses were conducted based on the methodology stated in section 1278 3.1 and presented in the following summary profiles based on individual 1279 pharmacological treatments (for full GRADE profiles, see appendix XXX).

1280 Antidepressants 1281 Table 58 Summary profile – amitriptyline as monotherapy (placebo- 1282 controlled trials) No of Relative risk Amitriptyline Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 4 to 8 weeks) 21 33/55 13/55 RR 2.54 (1.51 to 4.28) 36 more per 100 (from (60%) (23.6%) 12 more to 78 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 4 to 9 weeks) 42 58/111 37/115 RR 1.78 (0.76 to 4.21) 25 more per 100 (from (52.3%) (32.2%) 8 fewer to 100 more) LOW Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 9 weeks) 63 18/201 8/201 RR 2.03 (0.95 to 4.32) 4 more per 100 (from (9%) (4%) 0 fewer to 13 more) LOW

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 90 of 150

Primary outcome: dizziness (adverse effects) (follow-up 4 to 6 weeks) 24 2/31 3/32 RR 0.70 (0.13 to 3.73) 3 fewer per 100 (from (6.5%) (9.4%) 8 fewer to 26 more) LOW Primary outcome: dry mouth (adverse effects) (follow-up 4 to 6 weeks) 45 44/110 33/105 RR 1.30 (0.95 to 1.79) 9 more per 100 (from (40%) (31.4%) 2 fewer to 25 more) LOW Blurred vision (adverse effects) (follow-up 6 weeks) 26 4/62 5/59 RR 0.81 (0.24 to 2.79) 2 fewer per 100 (from (6.5%) (8.5%) 6 fewer to 15 more) LOW Primary outcome: gastrointestinal disturbances (adverse effects) (follow-up 6 weeks) 17 1/18 1/19 RR 1.06 (0.07 to 0 more per 100 (from VERY (5.6%) (5.3%) 15.64) 5 fewer to 77 more) LOW Primary outcome: sedation (adverse effects) (follow-up 4 weeks) 18 2/35 1/33 RR 1.89 (0.18 to 3 more per 100 (from VERY (5.7%) (3%) 19.83) 2 fewer to 57 more) LOW Primary outcome: Vomiting (adverse effects) (follow-up 6 weeks) 29 2/62 3/59 RR 0.82 (0.02 to 1 fewer per 100 (from (3.2%) (5.1%) 30.99) 5 fewer to 100 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 4 to 6 weeks)

310 78/92 49/88 RR 1.91 (0.6 to 6.09) 51 more per 100 (from (84.8%) (55.7%) 22 fewer to 100 more) LOW 1 Rintala et al. (2007); Vrethem et al. (1997). 2 Kautio et al. (2008); Kieburtz et al. (1998); Leijon et al. (1989); Vretham et al. (1997). 3 Cardenas et al. (2002); Kautio et al. (2008); Kieburtz et al. (1998); Rintala et al. (2007); Robinson et al. (2004); Vretham et al. (1997). 4 Kalso et al. (1996); Robinson et al. (2004). 5 Cardenas et al. (2002); Kalso et al. (1996); Robinson et al. (2004); Vretham et al. (1997). 6 Cardenas et al. (2002); Robinson et al. (2004). 7 Robinson et al. (2004). 8 Vretham et al. (1997). 9 Cardena et al. (2002); Robinson et al. (2004). 10 Cardenas et al. (2002); Leijon et al. (1989); Vretham et al. (1997). 1283

1284 Table 59 Summary profile – nortriptyline as monotherapy (placebo- 1285 controlled trials) No of Relative risk Nortriptyline Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 7 weeks) 11 12/31 11/33 RR 1.16 (0.6 to 2.24) 5 more per 100 (from (38.7%) (33.3%) 13 fewer to 41 more) MODERATE Primary outcome: dry mouth (adverse effects) (follow-up 7 weeks) 11 10/28 6/28 RR 1.67 (0.7 to 3.96) 14 more per 100 (from VERY (35.7%) (21.4%) 6 fewer to 63 more) LOW Blurred vision (adverse effects) (follow-up 7 weeks) 11 0/28 3/28 RR 0.14 (0.01 to 2.64) 9 fewer per 100 (from VERY (0%) (10.7%) 11 fewer to 18 more) LOW Primary outcome: gastrointestinal disturbances (adverse effects) (follow-up 7 weeks) 11 1/28 0/28 RR 3.00 (0.13 to – VERY (3.6%) (0%) 70.64) LOW Primary outcome: any adverse effects: unspecified (follow-up 7 weeks) 11 19/28 14/28 RR 1.36 (0.87 to 2.13) 18 more per 100 (from VERY (67.9%) (50%) 6 fewer to 57 more) LOW 1 Khoromi et al. (2007). 1286

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 91 of 150

1287 Table 60 Summary profile – imipramine as monotherapy (placebo- 1288 controlled trials) No of Relative risk Imipramine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 4 weeks) 11 14/33 2/33 RR 7.00 (1.72 to 36 more per 100 (from MODERATE (42.4%) (6.1%) 28.41) 4 more to 100 more) Primary outcome: number of withdrawals due to adverse effects (follow-up 4 weeks) 11 13/33 6/33 RR 2.17 (0.94 to 5.01) 21 more per 100 (from VERY (39.4%) (18.2%) 1 fewer to 73 more) LOW Primary outcome: dry mouth (adverse effects) (follow-up 4 weeks) 11 12/33 3/33 RR 4.00 (1.24 to 27 more per 100 (from VERY (36.4%) (9.1%) 12.88) 2 more to 100 more) LOW 1 Sindrup et al. (2003a) 1289

1290 Table 61 Summary profile – duloxetine as monotherapy (placebo- 1291 controlled trials) No of Relative risk Duloxetine Control Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 8 weeks) 11 11/24 4/24 RR 2.75 (1.02 to 7.44) 29 more per 100 (from (45.8%) (16.7%) 0 more to 100 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 8 weeks) 11 4/24 2/24 RR 2.00 (0.4 to 9.91) 8 more per 100 (from VERY (16.7%) (8.3%) 5 fewer to 74 more) LOW Primary outcome: dry mouth (adverse effects) (follow-up 8 weeks) 11 1/24 0/24 RR 3.00 (0.13 to – VERY (4.2%) (0%) 70.16) LOW 1 Vranken et al. (2011). 1292

1293 Table 63 Summary profile – venlafaxine as monotherapy (placebo- 1294 controlled trials) No of Relative risk Venlafaxine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 4 to 8 weeks) 21 28/69 10/52 RR 1.89 (0.65 to 5.52) 17 more per 100 (from (40.6%) (19.2%) 7 fewer to 87 more) LOW Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 8 weeks) 32 16/86 7/86 RR 2.07 (0.96 to 4.49) 9 more per 100 (from (18.6%) (8.1%) 0 fewer to 28 more) LOW Primary outcome: dry mouth (adverse effects) (follow-up 4 weeks) 23 12/46 9/46 RR 1.33 (0.68 to 2.62) 6 more per 100 (from (26.1%) (19.6%) 6 fewer to 32 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 weeks) 14 23/40 11/20 RR 1.05 (0.65 to 1.69) 3 more per 100 (from VERY (57.5%) (55%) 19 fewer to 38 more) LOW 1 Sindrup et al. (2003b); Yucel et al. (2004). 2 Sindrup et al. (2003b); Tasmuth et al. (2002); Yucel et al. (2004). 3 Sindrup et al. (2003b); Tasmuth et al. (2002). 4 Yucel et al. (2004). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 92 of 150

1295

1296 Anti-epileptics 1297 Table 63 Summary profile – gabapentin as monotherapy (placebo- 1298 controlled trials) No of Relative risk Gabapentin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 8 weeks) 11 5/22 6/22 RR 0.83 (0.3 to 2.33) 5 fewer per 100 (from (22.7%) (27.3%) 19 fewer to 36 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 8 weeks) 12 32/153 21/152 RR 1.51 (0.92 to 2.5) 7 more per 100 (from (20.9%) (13.8%) 1 fewer to 21 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 5 to 8 weeks) 33 92/263 41/259 RR 2.21 (1.6 to 3.06) 19 more per 100 (from (35%) (15.8%) 9 more to 33 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 8 weeks) 54 32/349 25/344 RR 1.26 (0.77 to 2.06) 2 more per 100 (from (9.2%) (7.3%) 2 fewer to 8 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 4 to 8 weeks) 55 89/398 28/391 RR 2.62 (1.58 to 4.36) 12 more per 100 (from (22.4%) (7.2%) 4 more to 24 more) MODERATE Primary outcome: somnolence (adverse effects) (follow-up 4 to 8 weeks) 36 41/187 12/182 RR 3.17 (1.74 to 5.8) 14 more per 100 (from (21.9%) (6.6%) 5 more to 32 more) LOW Primary outcome: sedation (adverse effects) (follow-up 8 weeks) 17 0/20 1/20 RR 0.33 (0.01 to 7.72) 3 fewer per 100 (from VERY (0%) (5%) 5 fewer to 34 more) LOW Primary outcome: fatigue (adverse effects) (follow-up 5 to 6 weeks) 28 32/211 19/209 RR 1.68 (1 to 2.82) 6 more per 100 (from (15.2%) (9.1%) 0 more to 17 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 4 to 8 weeks) 39 110/196 69/195 RR 1.58 (1.21 to 2.07) 21 more per 100 (from (56.1%) (35.4%) 7 more to 38 more) MODERATE 1 Rintala et al. (2007). 2 Serpell et al. (2002). 3 Gordh et al. (2008); Serpell et al. (2002); Smith et al. (2005). 4 Gordh et al. (20087; Hahn et al. (2004); Nikolajsen et al. (2006); Rintala et al. (2007); Serpell et al. (2002). 5 Bone et al. (2002); Gordh et al. (2008); Hahn et al. (2004); Rao et al. (2007); Serpell et al. (2002). 6 Bone et al. (2002); Hahn et al. (2004); Serpell et al. (2002). 7 Levendoglu et al. (2004). 8 Gordh et al. (2008); Rao et al. (2007). 9 Levendoglu et al. (2004); Nikolajsen et al. (2006); Serpell et al. (2002). 1299

1300 Table 64 Summary profile – pregabalin as monotherapy (placebo- 1301 controlled trials) No of Relative risk Pregabalin Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 12 to 14 weeks) 51 403/782 178/488 RR 1.44 (1.07 to 1.94) 16 more per 100 (from (51.5%) (36.5%) 3 more to 34 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 8 to 14 weeks) 42 253/655 96/361 RR 1.65 (0.94 to 2.89) 17 more per 100 (from

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 93 of 150

(38.6%) (26.6%) 2 fewer to 50 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 8 to 14 weeks) 43 448/713 255/421 RR 1.11 (0.85 to 1.45) 7 more per 100 (from (62.8%) (60.6%) 9 fewer to 27 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 14 weeks) 64 114/803 32/508 RR 2.07 (1.41 to 3.05) 7 more per 100 (from (14.2%) (6.3%) 3 more to 13 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 4 to 14 weeks) 55 165/606 43/430 RR 2.63 (1.52 to 4.54) 16 more per 100 (from (27.2%) (10%) 5 more to 35 more) LOW Primary outcome: somnolence (adverse effects) (follow-up 4 to 14 weeks) 55 123/606 36/430 RR 2.56 (1.32 to 4.96) 13 more per 100 (from (20.3%) (8.4%) 3 more to 33 more) LOW Primary outcome: fatigue (adverse effects) (follow-up 8 weeks) 16 15/127 10/127 RR 1.50 (0.7 to 3.21) 4 more per 100 (from VERY (11.8%) (7.9%) 2 fewer to 17 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 weeks) 27 180/289 97/205 RR 1.47 (1.27 to 1.71) 22 more per 100 (from (62.3%) (47.3%) 13 more to 34 more) MODERATE 1 Freynhagen et al. (2005); Moon et al. (2010); Siddall et al. (2006); Simpson et al. (2010); van Seventer et al. (2010). 2 Freynhagen et al. (2005); Moon et al. (2010); Siddall et al. (2006); Simpson et al. (2010). 3 Freynhagen et al. (2005); Moon et al. (2010); Simpson et al. (2010); van Seventer et al. (2010). 4 Freynhagen et al. (2005); Moon et al. (2010); Siddall et al. (2006); Simpson et al. (2010); van Seventer et al. (2010); Vranken et al. (2008). 5 Freynhagen et al. (2005); Siddall et al. (2006); Simpson et al. (2010); van Seventer et al. (2010); Vranken et al. (2008). 6 van Seventer et al. (2010). 7 Moon et al. (2010); van Seventer et al. (2010). 1302

1303 Table 65 Summary profile – lamotrigine as monotherapy (placebo- 1304 controlled trials) No of Relative risk Lamotrigine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 11 weeks) 11 5/11 2/11 RR 2.50 (0.61 to 27 more per 100 (from (45.5%) (18.2%) 10.25) 7 fewer to 100 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 8 to 14 weeks) 72 26/358 12/284 RR 1.81 (0.97 to 3.38) 3 more per 100 (from (7.3%) (4.2%) 0 fewer to 10 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 10 to 11 weeks) 23 2/78 2/77 RR 1.03 (0.16 to 6.81) 0 more per 100 (from (2.6%) (2.6%) 2 fewer to 15 more) LOW Primary outcome: fatigue (adverse effects) (follow-up 10 to 11 weeks) 23 4/78 4/77 RR 0.99 (0.27 to 3.68) 0 fewer per 100 (from (5.1%) (5.2%) 4 fewer to 14 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 8 to 9 weeks) 24 31/58 21/58 RR 1.95 (0.35 to 10.9) 34 more per 100 (from (53.4%) (36.2%) 24 fewer to 100 more) VERY LOW 1 Breuer et al. (2007). 2 Breuer et al. (2007); Finnerup et al. (2002); McCleane et al. (1999); Rao et al. (2008); Simpson et al. (2000); Simpson et al. (2003); Vestergaard et al. (2001). 3 Breuer et al. (2007); Rao et al. (2008). 4 Finnerup et al. (2002); Vestergaard et al. (2001). 1305

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 94 of 150

1306 Table 66 Summary profile – topiramate as monotherapy (placebo- 1307 controlled trials) No of Relative risk Topiramate Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 11 15/29 7/29 RR 2.14 (1.03 to 4.47) 28 more per 100 (from (51.7%) (24.1%) 1 more to 84 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 6 weeks) 11 10/41 0/41 RR 21.00 (1.27 to - VERY (24.4%) (0%) 364.93) LOW Primary outcome: fatigue (adverse effects) (follow-up 6 weeks) 11 10/29 9/29 RR 1.11 (0.53 to 2.33) 3 more per 100 (from VERY (34.5%) (31%) 15 fewer to 41 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 6 weeks) 11 25/29 21/29 RR 1.19 (0.91 to 1.56) 14 more per 100 (from VERY (86.2%) (72.4%) 7 fewer to 41 more) LOW 1 Khoromi et al. (2005). 1308

1309 Table 67 Summary profile – carbamazepine as monotherapy (placebo- 1310 controlled trials) No of Relative risk Carbamazepine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 1 to 46 months) 21 20/34 7/22 RR 1.88 (0.16 to 28 more per 100 (from (58.8%) (31.8%) 22.41) 27 fewer to 100 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 4 weeks) 12 13/15 7/15 RR 1.86 (1.04 to 3.30) 40 more per 100 (from VERY (86.7%) (46.7%) 2 more to 100 more) LOW 1 Leijon et al. (1989); Nicol et al. (1969). 2 Leijon et al. (1989). 1311 1312 Opioid analgesics 1313 Table 68 Summary profile – tramadol as monotherapy (placebo- 1314 controlled trials) No of Relative risk Tramadol Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 4 weeks) 11 4/12 0/12 RR 4.88 (0.28 to - (33.3%) (0%) 83.67) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 6 weeks) 32 21/86 4/75 RR 3.52 (1.37 to 9.03) 13 more per 100 (from (24.4%) (5.3%) 2 more to 43 more) LOW Primary outcome: constipation (adverse effects) (follow-up 4 to 6 weeks) 23 18/68 6/57 RR 2.09 (0.42 to 11 more per 100 (from (26.5%) (10.5%) 10.37) 6 fewer to 99 more) VERY LOW Primary outcome: nausea (adverse effects) (follow-up 4 to 6 weeks) 23 20/68 6/57 RR 2.47 (1.1 to 5.55) 15 more per 100 (from (29.4%) (10.5%) 1 more to 48 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 4 to 6 weeks) 23 27/68 5/57 RR 3.64 (1 to 13.21) 23 more per 100 (from Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 95 of 150

(39.7%) (8.8%) 0 more to 100 more) VERY LOW Primary outcome: any adverse effects (unspecified) (follow-up 4 to 6 weeks) 32 56/81 19/69 RR 2.07 (1.14 to 3.77) 29 more per 100 (from (69.1%) (27.5%) 4 more to 76 more) VERY LOW 1 Norrbrink et al. (2009). 2 Arbaiza et al. (2007); Norrbrink et al. (2009); Sindrup et al. (2003). 3 Norrbrink et al. (2009); Sindrup et al. (2003). 1315

1316 Table 69 Summary profile – morphine as monotherapy (placebo- 1317 controlled trials) No of Relative risk Morphine Placebo Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 7 weeks) 11 33/50 19/43 RR 1.49 (1.01 to 2.21) 22 more per 100 (from (66%) (44.2%) 0 more to 53 more) MODERATE Primary outcome: patient-reported 50% pain reduction (follow-up 4 to 7 weeks) 22 28/62 14/55 RR 1.75 (1.04 to 2.96) 19 more per 100 (from (45.2%) (25.5%) 1 more to 50 more) MODERATE Primary outcome: patient-reported global improvement/impression of change (follow-up 6 weeks) 13 13/32 11/33 RR 1.22 (0.64 to 2.31) 7 more per 100 (from (40.6%) (33.3%) 12 fewer to 44 more) MODERATE Primary outcome: number of withdrawals due to adverse effects (follow-up 6 weeks) 13 9/55 1/55 RR 9.00 (1.18 to 15 more per 100 (from VERY (16.4%) (1.8%) 68.66) 0 more to 100 more) LOW Primary outcome: constipation (adverse effects) (follow-up 6 weeks) 24 35/78 4/71 RR 8.12 (3.05 to 40 more per 100 (from VERY (44.9%) (5.6%) 21.61) 12 more to 100 more) LOW Primary outcome: somnolence/drowsiness (adverse effects) (follow-up 6 to 7 weeks) 24 16/78 4/71 RR 3.39 (1.17 to 9.76) 13 more per 100 (from (20.5%) (5.6%) 1 more to 49 more) LOW Primary outcome: nausea (adverse effects) (follow-up 6 to 7 weeks) 24 6/78 1/71 RR 3.94 (0.69 to 4 more per 100 (from (7.7%) (1.4%) 22.46) 0 fewer to 30 more) LOW Primary outcome: dizziness (adverse effects) (follow-up 6 to 7 weeks) 24 6/78 3/71 RR 1.86 (0.49 to 7.04) 4 more per 100 (from (7.7%) (4.2%) 2 fewer to 26 more) LOW Primary outcome: any adverse effects (unspecified) (follow-up 6 to 7 weeks) 24 53/78 21/71 RR 2.32 (1.26 to 4.28) 39 more per 100 (from (67.9%) (29.6%) 8 more to 97 more) VERY LOW 1 Wu et al. (2008). 2 Huse et al. (2001); Wu et al. (2008). 3 Khoromi et al. (2007). 4 Khoromi et al. (2007); Wu et al. (2008). 1318

1319 Topical treatments 1320 Table 70 Summary profile – topical capsaicin (0.075% cream) as 1321 monotherapy (placebo-controlled trials) No of Capsaicin Relative risk Control Absolute risk Quality studies 0.075% cream (95% CI) Primary outcome: patient-reported 50% pain reduction (follow-up 6 weeks) 11 8/13 3/9 RR 2.05 (0.73 to 5.80) 35 more per 100 (from (61.5%) (33.3%) 9 fewer to 100 more) MODERATE Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 96 of 150

Primary outcome: number of withdrawals due to adverse effects (follow-up 4 to 8 weeks) 32 24/167 5/161 RR 3.95 (1.65 to 9.42) 9 more per 100 (from (14.4%) (3.1%) 2 more to 26 more) LOW Primary outcome: burning (adverse effects) (follow-up 4 to 8 weeks) 43 155/225 62/231 RR 2.46 (1.33 to 4.58) 39 more per 100 (from (68.9%) (26.8%) 9 more to 96 more) LOW 1 Watson and Evans (1992). 2 Donofrio et al. (1991); Paice et al. (2000); Watson & Evans (1992). 3 Donofrio et al. (1991); Low et al. (1995); Paice et al. (2000); Watson & Evans (1992). 1322

1323 Table 71 Summary profile – topical lidocaine (5% patch/cream) as 1324 monotherapy (placebo-controlled trials) No of Lidocaine 5% Relative risk Control Absolute risk Quality studies patch/cream (95% CI) Primary outcome: withdrawals due to adverse effects (follow-up 1 week) 11 1/58 0/58 RR 3.00 (0.12 to – VERY (1.7%) (0%) 72.15) LOW Primary outcome: rash (adverse effects) (follow-up 1 week) 11 10/58 11/58 RR 0.91 (0.42 to 1.97) 2 fewer per 100 (from VERY (17.2%) (19.3%) 11 fewer to 19 more) LOW Primary outcome: skin irritation (adverse effects) (follow-up 1 week) 13 5/35 3/35 RR 1.67 (0.43 to 6.45) 6 more per 100 (from VERY (14.3%) (8.6%) 5 fewer to 47 more) LOW 1 Meier et al. (2003). 3 Ho et al. (2008). 1325

1326 Note: No study on topical lidocaine (5% patch/cream) as monotherapy that 1327 reported the critical outcomes on pain was identified or met the inclusion and 1328 exclusion criteria.

1329 No of Lidocaine 5% Relative risk Control Absolute risk Quality studies patch/cream (95% CI) Other reported pain outcome: pain intensity (scale: NRSpi 11-point) (follow-up 4 weeks) 11 8 13 Treatment = 4.4 (2.12) Placebo = 4.8 (1.71); p = 0.92 LOW Other reported pain outcome: pain relief (scale: Global Pain Relief Scale) (follow-up 2 weeks) 12 61 59 Treatment = 2.25 (5.94) Placebo = 2.23 (5.45); p = 0.715 LOW Other reported pain outcome: pain intensity (scale: VASpi-100mm) (follow-up 1 week) 13 30 31 Treatment = −5.7 (17.5) Placebo = −7.6 (23.9); p = 0.88 LOW Other reported pain outcome: pain intensity (Scale: VASpi-100mm) (follow-up 1 week) 14 40 40 Treatment = NR Placebo = NR; p = 0.002 LOW 1 Cheville et al. (2009). 2 Estanislao et al. (2004). 3 Ho et al. (2008). 4 Meier et al. (2003). 1330

1331 Head-to-head comparative trials (monotherapy) 1332 Table 72 Summary profile – amitriptyline vs gabapentin as monotherapy Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 97 of 150

1333 (comparative trials) No of Relative risk Amitriptyline Gabapentin Absolute risk Quality studies (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 8 weeks) 11 13/22 5/22 RR 2.60 (1.12 to 6.05) 36 more per 100 (from (59.1%) (22.7%) 3 more to 100 more) MODERATE Primary outcome: number of withdrawals owing to adverse effects (follow-up 8 weeks) 11 2/38 2/38 RR 1.00 (0.15 to 6.74) – VERY (5.3%) (5.3%) LOW 1 Rintala et al. (2007). 1334

1335 Table 73 Summary profile – amitriptyline vs carbamazepine as 1336 monotherapy (comparative trials) No of Relative risk Amitriptyline Carbamazepine Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 4 weeks) 11 10/15 5/14 RR 1.87 (0.85 to 31 more per 100 (from (66.7%) (35.7%) 4.11) 5 fewer to 100 more) MODERATE Primary outcome: any adverse effects: unspecified (follow-up 4 weeks) 11 14/15 13/14 RR 1.01 (0.82 to 1 more per 100 (from VERY (93.3%) (92.9%) 1.23) 17 fewer to 21 more) LOW 1 Leijon and Boivie (1989). 1337

1338 Table 74 Summary profile – pregabalin vs oxycodone as monotherapy 1339 (comparative trials) No of Relative risk Pregabalin Oxycodone Absolute risk Quality studies (95% CI) Primary outcome: number of withdrawals owing to adverse effects (follow-up 3 months) 11 9/134 11/106 RR 0.65 (0.28 to 1.50) 4 fewer per 100 (from VERY LOW (6.7%) (10.4%) 7 fewer to 5 more) Other pain outcome: mean pain intensity (NRS-11 point) (follow-up 3 months) 11 134 106 Pregabalin = decreased 46%; VERY LOW Oxycodone = decreased 76%; p <0.05 1 Gatti et al. (2009). 1340

1341 Note: no study on pregabalin vs oxycodone as monotherapy that reported the 1342 critical outcomes on pain was identified or met the inclusion and exclusion 1343 criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 98 of 150

1344 Table 75 Summary profile – pregabalin vs topical lidocaine (5% patch) as 1345 monotherapy (comparative trials) No of Lidocaine 5% Relative risk Pregabalin Absolute risk Quality studies patch (95% CI) Primary outcome: patient-reported 30% pain reduction (follow-up 4 weeks) 11 74/137 85/144 RR 0.92 (0.74 to 1.12) 5 fewer per 100 (from (54%) (59%) 15 fewer to 7 more) LOW Primary outcome: patient-reported 50% pain reduction (follow-up 4 weeks) 11 44/137 56/144 RR 0.83 (0.60 to 1.14) 7 fewer per 100 (from (32.1%) (38.9%) 16 fewer to 5 more) LOW Primary outcome: patient-reported global improvement/impression of change (follow-up 4 weeks) 11 65/137 72/144 RR 0.95 (0.75 to 1.21) 3 fewer per 100 (from (47.4%) (50%) 12 fewer to 11 more) LOW Primary outcome: number of withdrawals owing to adverse effects (follow-up 4 weeks) 11 36/153 4/155 RR 9.12 (3.33 to 25.0) 21 more per 100 (from VERY LOW (23.5%) (2.6%) 6 more to 62 more) Primary outcome: any adverse effects: unspecified (follow-up 4 weeks) 11 63/153 9/155 RR 7.09 (3.66 to 13.7) 35 more per 100 (from VERY LOW (41.2%) (5.8%) 15 more to 74 more) 1 Baron et al. (2009). 1346

1347 Table 76 Summary profile – imipramine vs venlafaxine as monotherapy 1348 (comparative trials) No of Relative risk Imipramine Venlafaxine Absolute risk Quality studies (95% CI) Primary outcome: patient-reported global improvement/impression of change (follow-up 4 weeks) 11 14/33 8/33 RR 1.75 (0.85 to 3.60) 18 more per 100 (from (42.4%) (24.2%) 4 fewer to 63 more) MODERATE Primary outcome: dizziness (adverse effects) (follow-up 4 weeks) 11 3/33 2/33 RR 1.50 (0.27 to 8.40) 3 more per 100 (from VERY (9.1%) (6.1%) 4 fewer to 45 more) LOW Primary outcome: dry mouth (adverse effects) (follow-up 4 weeks) 11 12/33 4/33 RR 3.00 (1.08 to 8.35) 24 more per 100 (from VERY (36.4%) (12.1%) 1 more to 89 more) LOW Primary outcome: any adverse effects: unspecified (follow-up 4 weeks) 11 13/33 11/33 RR 1.18 (0.62 to 2.25) 6 more per 100 (from VERY (39.4%) (33.3%) 13 fewer to 42 more) LOW 1 Sindrup et al. (2003). 1349 1350 Head-to-head comparative trials (combination therapy) 1351 Table 77 Summary profile – pregabalin + oxycodone as combination 1352 therapy vs pregabalin alone (comparative trials) No of Pregabalin + Relative risk Pregabalin Absolute risk Quality studies oxycodone (95% CI) Primary outcome: number of withdrawals owing to adverse effects (follow-up 3 months) 11 10/169 9/134 RR 0.88 (0.37 to 2.11) 1 fewer per 100 (from VERY LOW (5.9%) (6.7%) 4 fewer to 7 more) Other pain outcome: mean pain intensity (NRS-11 point) (follow-up 3 months) 11 169 134 Pregabalin + oxycodone = decreased 80% VERY LOW Pregabalin = decreased 46%; p <0.05 Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 99 of 150

1 Gatti et al. (2009). 1353

1354 Note: no study on pregabalin + oxycodone as combination therapy vs 1355 pregabalin alone that reported the critical outcomes on pain was identified or 1356 met the inclusion and exclusion criteria.

1357 Table 78 Summary profile – pregabalin + oxycodone as combination 1358 therapy vs oxycodone alone (comparative trials) No of Pregabalin + Relative risk Oxycodone Absolute risk Quality studies oxycodone (95% CI) Primary outcome: number of withdrawals owing to adverse effects (follow-up 3 months) 11 10/169 11/106 RR 0.51 (0.22 to 1.19) 5 fewer per 100 (from VERY LOW (5.9%) (10.4%) 8 fewer to 2 more) Other pain outcome: mean pain intensity (NRS-11 point) (follow-up 3 months) 11 169 106 Pregabalin + oxycodone = decreased 80% VERY LOW Oxycodone = decreased 76%; p >0.05 1 Gatti et al. (2009). 1359

1360 Note: No study on pregabalin + oxycodone as combination therapy vs 1361 oxycodone alone that reported the critical outcomes on pain was identified or 1362 met the inclusion and exclusion criteria.

1363 Table 79 Summary profile – gabapentin + nortriptyline as combination 1364 therapy vs gabapentin alone (comparative trials) No of Gabapentin + Relative Gabapentin Absolute Quality studies nortriptyline (95% CI) Other pain outcome: daily pain scores (numerical rating scale) (follow-up 6 weeks) 11 45 45 Combination lower than gabapentin = −0.9 (−1.4 to −0.3) LOW 1 Gilron et al. (2009). 1365

1366 Note: no study on gabapentin + nortriptyline as combination therapy vs 1367 gabapentin alone e that reported the critical outcomes on pain and adverse 1368 effects was identified or met the inclusion and exclusion criteria.

1369 3.4.3 Evidence statements

1370 For details of how the evidence is graded, see ‘The guidelines manual’.

1371 3.4.3.1 No study on clomipramine, desipramine, dosulepin (dothiepin), 1372 doxepin, lofepramine, trimipramine, citalopram, fluoxetine, Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 100 of 150

1373 paroxetine, sertraline, oxcarbazepine, phenytoin, sodium valproate, 1374 buprenorphine, co-codamol, codeine phosphate, co-dydramol, 1375 dihydrocodeine, fentanyl was identified or met the inclusion and 1376 exclusion criteria for other neuropathic pain conditions (apart from 1377 PDN and PHN).

1378 Antidepressants as monotherapy against placebo 1379 Amitriptyline (linked to table 58) 1380 Critical outcomes (pain) 1381 3.4.3.2 Moderate quality evidence from two studies with 110 patients with 1382 other neuropathic pain (apart from PDN and PHN), showed that 1383 amitriptyline is more effective than placebo in achieving at least 1384 30% pain reduction from baseline up to 8 weeks’ follow-up.

1385 3.4.3.3 Low quality evidence from three studies with 226 patients with 1386 other neuropathic pain (apart from PDN and PHN), showed that 1387 there is no significant difference between patients on amitriptyline 1388 and placebo in achieving global improvement/impression of change 1389 from baseline up to 9 weeks’ follow-up.

1390 Critical outcomes (adverse effects) 1391 3.4.3.4 Low quality evidence from six studies with 402 patients with other 1392 neuropathic pain (apart from PDN and PHN), showed that there is 1393 no significant difference between patients on amitriptyline and 1394 placebo withdrawing from studies due to adverse effects from 1395 baseline up to 9 weeks’ follow-up.

1396 3.4.3.5 Low quality evidence from two studies with 63 patients with other 1397 neuropathic pain (apart from PDN and PHN), showed that there is 1398 no significant difference between patients on amitriptyline and 1399 placebo in experiencing dizziness from baseline up to 6 weeks’ 1400 follow-up.

1401 3.4.3.6 Low quality evidence from four studies with 215 patients with other 1402 neuropathic pain (apart from PDN and PHN), showed that there is 1403 no significant difference between patients on amitriptyline and Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 101 of 150

1404 placebo in experiencing dry mouth from baseline up to 6 weeks’ 1405 follow-up.

1406 3.4.3.7 Low quality evidence from two studies with 121 patients with other 1407 neuropathic pain (apart from PDN and PHN), showed that there is 1408 no significant difference between patients on amitriptyline and 1409 placebo in experiencing blurred vision and vomiting from baseline 1410 up to 6 weeks’ follow-up.

1411 3.4.3.8 Very low quality evidence from one study with 37 patients with 1412 other neuropathic pain (apart from PDN and PHN), showed that 1413 there is no significant difference between patients on amitriptyline 1414 and placebo in experiencing gastrointestinal disturbances from 1415 baseline up to 6 weeks’ follow-up.

1416 3.4.3.9 Very low quality evidence from one study with 68 patients with 1417 other neuropathic pain (apart from PDN and PHN), showed that 1418 there is no significant difference between patients on amitriptyline 1419 and placebo in experiencing sedation from baseline up to 4 weeks’ 1420 follow-up.

1421 3.4.3.10 Low quality evidence from three studies with 180 patients with 1422 other neuropathic pain (apart from PDN and PHN), showed that 1423 there is no significant difference between patients on amitriptyline 1424 and placebo in experiencing any adverse effects (unspecified) from 1425 baseline up to 6 weeks’ follow-up.

1426 Nortriptyline (linked to table 59) 1427 Critical outcomes (pain) 1428 3.4.3.11 Moderate quality evidence from one study with 64 patients with 1429 other neuropathic pain (apart from PDN and PHN), showed that 1430 there is no significant difference between nortriptyline and placebo 1431 in achieving global improvement/impression of change from 1432 baseline up to 7 weeks follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 102 of 150

1433 Critical outcomes (adverse effects) 1434 3.4.3.12 Very low quality evidence from one study with 56 patients with 1435 other neuropathic pain (apart from PDN and PHN), showed that 1436 there is no significant difference between patients on nortriptyline 1437 and placebo in experiencing dry mouth, blurred vision, 1438 gastrointestinal disturbances and any adverse effects (unspecified) 1439 from baseline up to 7 weeks’ follow-up.

1440 Imipramine (linked to table 60) 1441 Critical outcomes (pain) 1442 3.4.3.13 Moderate quality evidence from one study with 66 patients with 1443 other neuropathic pain (apart from PDN and PHN), showed that 1444 imipramine is more effective than placebo in achieving global 1445 improvement/impression of change from baseline up to 4 weeks’ 1446 follow-up.

1447 Critical outcomes (adverse effects) 1448 3.4.3.14 Very low quality evidence from one study with 66 patients with 1449 other neuropathic pain (apart from PDN and PHN), showed that 1450 there is no significant difference between patients on imipramine 1451 and placebo withdrawing from studies due to adverse effects from 1452 baseline up to 4 weeks’ follow-up.

1453 3.4.3.15 Very low quality evidence from one study with 66 patients with 1454 other neuropathic pain (apart from PDN and PHN), showed that 1455 patients on imipramine are more likely to experience dry mouth 1456 from baseline up to 4 weeks’ follow-up.

1457 Duloxetine (linked to table 61) 1458 Critical outcomes (pain) 1459 3.4.3.16 Moderate quality evidence from one study with 48 patients with 1460 other neuropathic pain (apart from PDN and PHN), showed that 1461 duloxetine is more effective than placebo in achieving global 1462 improvement/impression of change from baseline up to 8 weeks’ 1463 follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 103 of 150

1464 Critical outcomes (adverse effects) 1465 3.4.3.17 Very low quality evidence from one study with 48 patients with 1466 other neuropathic pain (apart from PDN and PHN), showed that 1467 there is no significant difference between patients on duloxetine 1468 and placebo in experiencing dizziness and dry mouth from baseline 1469 up to 8 weeks’ follow-up.

1470 Venlafaxine (linked to table 62) 1471 Critical outcomes (pain) 1472 3.4.3.18 Low quality evidence from two studies with 121 patients with other 1473 neuropathic pain (apart from PDN and PHN), showed that there is 1474 no significant difference between venlafaxine and placebo in 1475 achieving global improvement/impression of change from baseline 1476 up to 8 weeks follow-up.

1477 Critical outcomes (adverse effects) 1478 3.4.3.19 Low quality evidence from three studies with 172 patients with 1479 other neuropathic pain (apart from PDN and PHN), showed that 1480 there is no significant difference between patients on venlafaxine 1481 and placebo withdrawing from studies due to adverse effects from 1482 baseline up to 8 weeks’ follow-up.

1483 3.4.3.20 Low quality evidence from two studies with 92 patients with other 1484 neuropathic pain (apart from PDN and PHN), showed that there is 1485 no significant difference between patients on venlafaxine and 1486 placebo in experiencing dry mouth from baseline up to 4 weeks’ 1487 follow-up.

1488 3.4.3.21 Very low quality evidence from one study with 60 patients with 1489 other neuropathic pain (apart from PDN and PHN), showed that 1490 there is no significant difference between patients on venlafaxine 1491 and placebo in experiencing any adverse effects (unspecified) from 1492 baseline up to 8 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 104 of 150

1493 Anti-epileptics as monotherapy against placebo 1494 Gabapentin (linked to table 63) 1495 Critical outcomes (pain) 1496 3.4.3.22 Moderate quality evidence from one study with 44 patients, and 1497 one study with 305 patients with other neuropathic pain (apart from 1498 PDN and PHN), showed that there is no significant difference 1499 between gabapentin and placebo in achieving at least 30% and 1500 50% pain reduction respectively from baseline up to 8 weeks’ 1501 follow-up.

1502 3.4.3.23 Moderate quality evidence from three studies with 522 patients with 1503 other neuropathic pain (apart from PDN and PHN), showed that 1504 gabapentin is more effective than placebo in achieving global 1505 improvement/impression of change from baseline up to 8 weeks’ 1506 follow-up.

1507 Critical outcomes (adverse effects) 1508 3.4.3.24 Low quality evidence from five studies with 693 patients with other 1509 neuropathic pain (apart from PDN and PHN), showed that there is 1510 no significant difference between patients on gabapentin and 1511 placebo withdrawing from studies due to adverse effects from 1512 baseline up to 8 weeks’ follow-up.

1513 3.4.3.25 Moderate quality evidence from five studies with 789 patients with 1514 other neuropathic pain (apart from PDN and PHN), showed that 1515 patients on gabapentin are more likely to experience dizziness 1516 compared with placebo from baseline up to 8 weeks’ follow-up.

1517 3.4.3.26 Low quality evidence from three studies with 369 patients with 1518 other neuropathic pain (apart from PDN and PHN), showed that 1519 patients on gabapentin are more likely to experience somnolence 1520 compared with placebo from baseline up to 8 weeks’ follow-up.

1521 3.4.3.27 Low quality evidence from two studies with 420 patients with other 1522 neuropathic pain (apart from PDN and PHN), showed that patients

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 105 of 150

1523 on gabapentin are more likely to experience fatigue compared with 1524 placebo from baseline up to 6 weeks’ follow-up.

1525 3.4.3.28 Very low quality evidence from one study with 40 patients with 1526 other neuropathic pain (apart from PDN and PHN), showed that 1527 there is no significant difference between patients on gabapentin 1528 and placebo in experiencing sedation from baseline up to 8 weeks’ 1529 follow-up.

1530 3.4.3.29 Moderate quality evidence from three studies with 391 patients with 1531 other neuropathic pain (apart from PDN and PHN), showed that 1532 patients on gabapentin are more likely to experience any adverse 1533 effects (unspecified) compared with placebo from baseline up to 1534 8 weeks’ follow-up.

1535 Pregabalin (linked to table 64) 1536 Critical outcomes (pain) 1537 3.4.3.30 Moderate quality evidence from five studies with 1270 patients with 1538 other neuropathic pain (apart from PDN and PHN), showed that 1539 pregabalin is more effective than placebo in achieving at least 30% 1540 pain reduction from baseline up to 14 weeks’ follow-up.

1541 3.4.3.31 Moderate quality evidence from four studies with 1016 patients, 1542 and four studies with 1134 patients with other neuropathic pain 1543 (apart from PDN and PHN), showed that there is no significant 1544 difference between pregabalin and placebo in achieving at least 1545 50% pain reduction and global improvement/impression of change 1546 respectively from baseline up to 14 weeks’ follow-up.

1547 Critical outcomes (adverse effects) 1548 3.4.3.32 Moderate quality evidence from six studies with 1311 patients with 1549 other neuropathic pain (apart from PDN and PHN), showed that 1550 patients on pregabalin are more likely withdraw from studies due to 1551 adverse effects compared with placebo from baseline up to 1552 14 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 106 of 150

1553 3.4.3.33 Low quality evidence from five studies with 1036 patients with other 1554 neuropathic pain (apart from PDN and PHN), showed that patients 1555 on pregabalin are more likely to experience dizziness and 1556 somnolence compared with placebo from baseline up to 14 weeks’ 1557 follow-up.

1558 3.4.3.34 Very low quality evidence from one study with 154 patients with 1559 other neuropathic pain (apart from PDN and PHN), showed that 1560 there is no significant difference between patients on pregabalin 1561 and placebo in experiencing fatigue from baseline up to 8 weeks’ 1562 follow-up.

1563 3.4.3.35 Moderate quality evidence from two studies with 494 patients with 1564 other neuropathic pain (apart from PDN and PHN), showed that 1565 patients on pregabalin are more likely to experience any adverse 1566 effects (unspecified) compared with placebo from baseline up to 1567 8 weeks’ follow-up.

1568 Lamotrigine (linked to table 65) 1569 Critical outcomes (pain) 1570 3.4.3.36 Moderate quality evidence from one study with 22 patients with 1571 other neuropathic pain (apart from PDN and PHN), showed that 1572 there is no significant difference between lamotrigine and placebo 1573 in achieving at least 30% pain reduction from baseline up to 1574 11 weeks’ follow-up.

1575 Critical outcomes (adverse effects) 1576 3.4.3.37 Low quality evidence from seven studies with 642 patients with 1577 other neuropathic pain (apart from PDN and PHN), showed that 1578 there is no significant difference between patients on lamotrigine 1579 and placebo withdrawing from studies due to adverse effects from 1580 baseline up to 14 weeks’ follow-up.

1581 3.4.3.38 Low quality evidence from two studies with 155 patients with other 1582 neuropathic pain (apart from PDN and PHN), showed that there is 1583 no significant difference between patients on lamotrigine and Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 107 of 150

1584 placebo in experiencing dizziness and fatigue from baseline up to 1585 11 weeks’ follow-up.

1586 3.4.3.39 Very low quality evidence from two studies with 116 patients with 1587 other neuropathic pain (apart from PDN and PHN), showed that 1588 there is no significant difference between patients on lamotrigine 1589 and placebo in experiencing any adverse effects (unspecified) from 1590 baseline up to 9 weeks’ follow-up.

1591 Topiramate (linked to table 66) 1592 Critical outcomes (pain) 1593 3.4.3.40 Moderate quality evidence from one study with 58 patients with 1594 other neuropathic pain (apart from PDN and PHN), showed that 1595 topiramate is more effective than placebo in achieving global 1596 improvement/impression of change from baseline up to 6 weeks’ 1597 follow-up.

1598 Critical outcomes (adverse effects) 1599 3.4.3.41 Very low quality evidence from one study with 82 patients with 1600 other neuropathic pain (apart from PDN and PHN), showed that 1601 patients on topiramate are more likely to withdraw from studies due 1602 to adverse effects compared with placebo from baseline up to 1603 6 weeks’ follow-up.

1604 3.4.3.42 Very low quality evidence from one study with 58 patients with 1605 other neuropathic pain (apart from PDN and PHN), showed that 1606 there is no significant difference between patients on topiramate 1607 and placebo in experiencing fatigue and any adverse effects 1608 (unspecified) from baseline up to 6 weeks’ follow-up.

1609 Carbamazepine (linked to table 67) 1610 Critical outcomes (pain) 1611 3.4.3.43 Low quality evidence from two studies with 56 patients with other 1612 neuropathic pain (apart from PDN and PHN), showed that there is 1613 no significant difference between carbamazepine and placebo in

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 108 of 150

1614 achieving global improvement/impression of change from baseline 1615 up to 46 months’ follow-up.

1616 Critical outcomes (adverse effects) 1617 3.4.3.44 Very low quality evidence from one study with 30 patients with 1618 other neuropathic pain (apart from PDN and PHN), showed that 1619 patients on carbamazepine are more likely to experience any 1620 adverse effects (unspecified) compared with placebo from baseline 1621 up to 4 weeks’ follow-up.

1622 Opioid analgesics as monotherapy against placebo 1623 Tramadol (linked to table 68) 1624 Critical outcomes (pain) 1625 3.4.3.45 Moderate quality evidence from one study with 24 patients with 1626 other neuropathic pain (apart from PDN and PHN), showed that 1627 there is no significant difference between tramadol and placebo in 1628 achieving global improvement/impression of change from baseline 1629 up to 4 weeks’ follow-up.

1630 Critical outcomes (adverse effects) 1631 3.4.3.46 Low quality evidence from three studies with 161 patients with 1632 other neuropathic pain (apart from PDN and PHN), showed that 1633 patients on tramadol are more likely to withdraw from studies due 1634 to adverse effects compared with placebo from baseline up to 1635 6 weeks’ follow-up.

1636 3.4.3.47 Low quality evidence from two studies with 125 patients with other 1637 neuropathic pain (apart from PDN and PHN), showed that patients 1638 on tramadol are more likely to experience nausea compared with 1639 placebo from baseline up to 6 weeks’ follow-up.

1640 3.4.3.48 Very low quality evidence from two studies with 125 patients with 1641 other neuropathic pain (apart from PDN and PHN), showed that 1642 patients on tramadol are more likely to experience dizziness 1643 compared with placebo from baseline up to 6 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 109 of 150

1644 3.4.3.49 Very low quality evidence from two studies with 125 patients with 1645 other neuropathic pain (apart from PDN and PHN), showed that 1646 there is no significant difference between patients on tramadol and 1647 placebo in experiencing constipation from baseline up to 6 weeks’ 1648 follow-up.

1649 3.4.3.50 Very low quality evidence from three studies with 150 patients with 1650 other neuropathic pain (apart from PDN and PHN), showed that 1651 patients on tramadol are more likely to experience any adverse 1652 effects (unspecified) compared with placebo from baseline up to 6 1653 weeks’ follow-up.

1654 Morphine (linked to table 69) 1655 Critical outcomes (pain) 1656 3.4.3.51 Moderate quality evidence from one study with 93 patients, two 1657 studies with 117 patients and one study with 65 patients with other 1658 neuropathic pain (apart from PDN and PHN), showed that 1659 morphine is more effective than placebo in achieving at least 30% 1660 or 50% pain reduction and global improvement/impression of 1661 change respectively from baseline up to 7 weeks’ follow-up.

1662 Critical outcomes (adverse effects) 1663 3.4.3.52 Very low quality evidence from one study with 110 patients with 1664 other neuropathic pain (apart from PDN and PHN), showed that 1665 patients on morphine are more likely to withdraw from studies due 1666 to adverse effects compared with placebo from baseline up to 1667 6 weeks’ follow-up.

1668 3.4.3.53 Very low quality evidence from two studies with 149 patients with 1669 other neuropathic pain (apart from PDN and PHN), showed that 1670 patients on morphine are more likely to experience constipation 1671 and any adverse effects (unspecified) compared with placebo from 1672 baseline up to 7 weeks’ follow-up.

1673 3.4.3.54 Low quality evidence from two studies with 149 patients with other 1674 neuropathic pain (apart from PDN and PHN), showed that patients Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 110 of 150

1675 on morphine are more likely to experience somnolence compared 1676 with placebo from baseline up to 7 weeks’ follow-up.

1677 3.4.3.55 Low quality evidence from two studies with 149 patients with other 1678 neuropathic pain (apart from PDN and PHN), showed that there is 1679 no significant difference between patients on morphine and placebo 1680 in experiencing nausea and dizziness from baseline up to 7 weeks’ 1681 follow-up.

1682 Topical treatments as monotherapy against placebo 1683 Topical capsaicin (0.075% cream) (linked to table 70) 1684 Critical outcomes (pain) 1685 3.4.3.56 Moderate quality evidence from one study with 22 patients with 1686 other neuropathic pain (apart from PDN and PHN), showed that 1687 there is no significant difference between topical capsaicin (0.075% 1688 cream) and placebo in achieving at least 50% pain reduction from 1689 baseline up to 6 weeks’ follow-up.

1690 Critical outcomes (adverse effects) 1691 3.4.3.57 Low quality evidence from three studies with 328 patients and four 1692 studies with 456 patients with other neuropathic pain (apart from 1693 PDN and PHN), showed that patients on between topical capsaicin 1694 (0.075% cream) are more likely to withdraw from studies due to 1695 adverse effects and burning compared with placebo from baseline 1696 up to 8 weeks’ follow-up.

1697 Topical lidocaine (5% patch/cream) (linked to table 71) 1698 Critical outcomes (pain) 1699 3.4.3.58 No study on topical lidocaine (5% patch/cream) as monotherapy 1700 that reported the critical outcomes on pain was identified or met the 1701 inclusion and exclusion criteria.

1702 Critical outcomes (adverse effects) 1703 3.4.3.59 Very low quality evidence from one study with 116 patients with 1704 other neuropathic pain (apart from PDN and PHN), showed that 1705 there is no significant difference between patients on topical Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 111 of 150

1706 lidocaine (5% patch/cream) and placebo withdrawing from studies 1707 due to adverse effects, or experiencing skin rash from baseline up 1708 to 1 week follow-up.

1709 3.4.3.60 Very low quality evidence from one study with 70 patients with 1710 other neuropathic pain (apart from PDN and PHN), showed that 1711 there is no significant difference between patients on topical 1712 lidocaine (5% patch/cream) and placebo in experiencing skin 1713 irritation from baseline up to 1 week follow-up.

1714 Other reported pain outcomes 1715 3.4.3.61 Low quality evidence from one study with 21 patients and one 1716 study with 71 patients with other neuropathic pain (apart from PDN 1717 and PHN), showed that there is no significant difference between 1718 patients on topical lidocaine (5% patch/cream) and placebo in pain 1719 intensity scores from baseline up to 4 weeks’ follow-up.

1720 3.4.3.62 Low quality evidence from one study with 120 patients with other 1721 neuropathic pain (apart from PDN and PHN), showed that there is 1722 no significant difference between patients on topical lidocaine (5% 1723 patch/cream) and placebo in pain relief scores from baseline up to 1724 4 weeks’ follow-up.

1725 3.4.3.63 Low quality evidence from one study with 80 patients with other 1726 neuropathic pain (apart from PDN and PHN), showed that patients 1727 on topical lidocaine (5% patch/cream) have better pain relief scores 1728 compared with placebo from baseline up to 1 week follow-up.

1729 Head-to-head comparative trials (monotherapy) 1730 Amitriptyline vs gabapentin (linked to table 72) 1731 Critical outcomes (pain) 1732 3.4.3.64 Moderate quality evidence from one study with 44 patients with 1733 other neuropathic pain (apart from PDN and PHN), showed that 1734 amitriptyline is more effective than gabapentin in achieving at least 1735 30% pain reduction from baseline up to 8 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 112 of 150

1736 Critical outcomes (adverse effects) 1737 3.4.3.65 Very low quality evidence from one study with 76 patients with 1738 other neuropathic pain (apart from PDN and PHN), showed that 1739 there is no significant difference between patients on amitriptyline 1740 and patients on gabapentin withdrawing from studies due to 1741 adverse effects from baseline up to 8 weeks’ follow-up.

1742 Amitriptyline vs carbamazepine (linked to table 73) 1743 Critical outcomes (pain) 1744 3.4.3.66 Moderate quality evidence from one study with 29 patients with 1745 other neuropathic pain (apart from PDN and PHN), showed that 1746 there is no significant difference between amitriptyline and 1747 carbamazepine in achieving global improvement/impression of 1748 change from baseline up to 4 weeks’ follow-up.

1749 Critical outcomes (adverse effects) 1750 3.4.3.67 Very low quality evidence from one study with 29 patients with 1751 other neuropathic pain (apart from PDN and PHN), showed that 1752 there is no significant difference between patients on amitriptyline 1753 and patients on carbamazepine in experiencing any adverse effects 1754 (unspecified) from baseline up to 4 weeks’ follow-up.

1755 Pregabalin vs oxycodone (linked to table 74) 1756 Critical outcomes (pain) 1757 3.4.3.68 No study on pregabalin vs oxycodone as monotherapy that 1758 reported the critical outcomes on pain was identified or met the 1759 inclusion and exclusion criteria.

1760 Critical outcomes (adverse effects) 1761 3.4.3.69 Very low quality evidence from one study with 240 patients with 1762 other neuropathic pain (apart from PDN and PHN), showed that 1763 there is no significant difference between patients on pregabalin 1764 and patients on oxycodone withdrawing from studies due to 1765 adverse effects from baseline up to 12 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 113 of 150

1766 Other reported pain outcomes 1767 3.4.3.70 Very low quality evidence from one study with 140 patients with 1768 other neuropathic pain (apart from PDN and PHN), showed that 1769 patients on oxycodone have better pain relief scores compared with 1770 patients on pregabalin from baseline up to 12 weeks’ follow-up.

1771 Pregabalin vs topical lidocaine (5% patch) (linked to table 75) 1772 Critical outcomes (pain) 1773 3.4.3.71 Low quality evidence from one study with 281 patients with other 1774 neuropathic pain (apart from PDN and PHN), showed that there is 1775 no significant difference between pregabalin and topical lidocaine 1776 (5% patch) in achieving at least 30% or 50% pain reduction and 1777 global improvement/impression of change from baseline up to 1778 4 weeks’ follow-up.

1779 Critical outcomes (adverse effects) 1780 3.4.3.72 Very low quality evidence from one study with 308 patients with 1781 other neuropathic pain (apart from PDN and PHN), showed that 1782 patients on pregabalin are more likely to withdraw from studies due 1783 to adverse effects, and experience any adverse effects 1784 (unspecified) compared with topical lidocaine (5% patch) from 1785 baseline up to 4 weeks’ follow-up.

1786 Imipramine vs venlafaxine (linked to table 76) 1787 Critical outcomes (pain) 1788 3.4.3.73 Moderate quality evidence from one study with 66 patients with 1789 other neuropathic pain (apart from PDN and PHN), showed that 1790 there is no significant difference between imipramine and 1791 venlafaxine in achieving global improvement/impression of change 1792 from baseline up to 4 weeks’ follow-up.

1793 Critical outcomes (adverse effects) 1794 3.4.3.74 Very low quality evidence from one study with 66 patients with 1795 other neuropathic pain (apart from PDN and PHN), showed that 1796 there is no significant difference between patients on imipramine

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 114 of 150

1797 and venlafaxine in experiencing dizziness and any adverse effects 1798 (unspecified) from baseline up to 4 weeks’ follow-up.

1799 3.4.3.75 Very low quality evidence from one study with 66 patients with 1800 other neuropathic pain (apart from PDN and PHN), showed that 1801 patients on imipramine are more likely to experience dry mouth 1802 compared with venlafaxine from baseline up to 4 weeks’ follow-up.

1803 Head-to-head comparative trials (combination therapy) 1804 Pregabalin + oxycodone as combination therapy vs pregabalin alone 1805 (linked to table 77) 1806 Critical outcomes (pain) 1807 3.4.3.76 No study on pregabalin + oxycodone as combination therapy vs 1808 pregabalin alone that reported the critical outcomes on pain was 1809 identified or met the inclusion and exclusion criteria.

1810 Critical outcomes (adverse effects) 1811 3.4.3.77 Very low quality evidence from one study with 303 patients with 1812 other neuropathic pain (apart from PDN and PHN), showed that 1813 there is no significant difference between patients on pregabalin + 1814 oxycodone and patients on pregabalin alone withdrawing from 1815 studies due to adverse effects from baseline up to 12 weeks’ 1816 follow-up.

1817 Other reported pain outcomes 1818 3.4.3.78 Very low quality evidence from one study with 303 patients with 1819 other neuropathic pain (apart from PDN and PHN), showed that 1820 patients on pregabalin + oxycodone have better pain intensity 1821 scores compared with pregabalin alone from baseline up to 1822 12 weeks’ follow-up.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 115 of 150

1823 Pregabalin + oxycodone as combination therapy vs oxycodone alone 1824 (linked to table 78) 1825 Critical outcomes (pain) 1826 3.4.3.79 No study on pregabalin + oxycodone as combination therapy vs 1827 oxycodone alone that reported the critical outcomes on pain was 1828 identified or met the inclusion and exclusion criteria.

1829 Critical outcomes (adverse effects) 1830 3.4.3.80 Very low quality evidence from one study with 275 patients with 1831 other neuropathic pain (apart from PDN and PHN), showed that 1832 there is no significant difference between patients on pregabalin + 1833 oxycodone and patients on oxycodone alone withdrawing from 1834 studies due to adverse effects from baseline up to 12 weeks’ 1835 follow-up.

1836 Other reported pain outcomes 1837 3.4.3.81 Very low quality evidence from one study with 275 patients with 1838 other neuropathic pain (apart from PDN and PHN), showed that 1839 there is no significant difference in pain intensity scores between 1840 patients on pregabalin + oxycodone and patients on oxycodone 1841 alone from baseline up to 12 weeks’ follow-up.

1842 Gabapentin + nortriptyline as combination therapy vs gabapentin alone 1843 (linked to table 79) 1844 Critical outcomes (pain) 1845 3.4.3.82 No studies on gabapentin + nortriptyline as combination therapy vs 1846 gabapentin alone that reported the critical outcomes on pain were 1847 identified or met the inclusion and exclusion criteria.

1848 Critical outcomes (adverse effects) 1849 3.4.3.83 No studies on gabapentin + nortriptyline as combination therapy vs 1850 gabapentin alone that reported the critical outcomes on adverse 1851 effects were identified or met the inclusion and exclusion criteria.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 116 of 150

1852 Other reported pain outcomes 1853 3.4.3.84 Low quality evidence from one study with 90 patients with other 1854 neuropathic pain (apart from PDN and PHN), showed that there is 1855 no significant difference in daily pain scores between patients on 1856 gabapentin + nortriptyline and patients on gabapentin alone from 1857 baseline up to 6 weeks’ follow-up.

1858 3.4.4 Health economic modelling

1859 No appropriate economic analyses were identified in the literature to inform 1860 recommendations for other neuropathic conditions other than PDN and PHN. 1861 The GDG considered that it would be appropriate to extrapolate results from 1862 the draft HTA reports analysis of PHN and PDN with consideration given to 1863 the findings of the clinical review. However, the GDG noted that for both 1864 models the QALY differences were negligible and that costs were the main 1865 contributor to difference in cost effectiveness. Therefore, given this 1866 information the GDG felt confident that it could extrapolate the results.

1867 Health economics evidence statements – neuropathic pain excluding 1868 PDN and PHN 1869 3.4.4.1 No cost effectiveness or economic study comparing treatments for 1870 neuropathic pain excluding PHN and PDN was identified or met the 1871 inclusion and exclusion criteria.

1872 3.4.5 Evidence to recommendations

1873 As stated in section 3.3.5, this section will consider both the evidence-base for 1874 post-herpetic neuralgia (PHN) and for other neuropathic pain conditions (that 1875 is all other neuropathic pain conditions apart from PDN and PHN).

Relative value of The relative value of different outcomes considered by the GDG for PHN different outcomes and other neuropathic pain conditions are the same as in PDN (please see section 3.2.5).

Quality of evidence The GDG agreed that when discussing the quality of evidence, considerations of the number of studies, the size of the study population and the magnitude of effects are important.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 117 of 150

Overall, the GDG agreed that the core evidence-base is from placebo- controlled trials, and evidence on head-to-head comparative trials and trials on combination therapy is very limited. Hence, the GDG felt that it could not confidently draw conclusions solely based on this evidence. The focus of the discussion is based on the placebo-controlled trials and evidence from health economics evaluation.

Antidepressants For post-herpetic neuralgia (PHN) The GDG agreed that there is moderate-quality evidence on the statistically significant efficacy of amitriptyline for 30% pain reduction, and desipramine for global improvement/impression of change.

For other neuropathic pain conditions ( that is all other neuropathic pain conditions apart from PDN and PHN) The GDG agreed that there is also moderate-quality evidence on the statistically significant efficacy of amitriptyline in 30% pain reduction, and of imipramine and duloxetine in global improvement/impression of change.

The GDG agreed that there was insufficient evidence on the efficacy of nortriptyline and venlafaxine.

Overall Amitriptyline seems to have moderate-quality evidence on its efficacy for both PHN and other neuropathic pain conditions (apart from PDN) in 30% pain reduction.

Anti-epileptics For post-herpetic neuralgia (PHN) The GDG agreed that there is moderate-quality evidence on the statistically significant efficacy of gabapentin in 30% pain reduction, and high/moderate-quality evidence on the significant efficacy of pregabalin in 30%, 50% pain reduction and global improvement/impression of change.

For other neuropathic pain conditions ( that is all other neuropthic pain conditions apart from PDN and PHN) The GDG agreed that there is also moderate-quality evidence on the statistically significant efficacy of gabapentin in global improvement/impression of change, pregabalin in 30% pain reduction, and topiramate in global improvement/impression of change. However, the moderate-quality evidence on topiramate is based on a single study with small sample size (less than 100) by comparison with gabapentin (nine

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 118 of 150

studies) and pregabalin (six studies). Hence, the GDG was concerned that the effect size for topiramate in this single study may be an overestimate.

The GDG agreed that there was insufficient evidence on the efficacy of lamotrigine and carbamazepine.

Overall Gabapentin and pregabalin seem to have high to moderate quality evidence on efficacy for both PHN and other neuropathic pain conditions ( that is all other neuropthic pain conditions apart from PDN and PHN) in all critical outcomes on pain.

Opioid analgesics For post-herpetic neuralgia (PHN) The GDG agreed that there is moderate-quality evidence on the significant efficacy of tramadol in 50% pain reduction for PHN.

For other neuropathic pain conditions (other than PDN and PHN) The GDG agreed that there is moderate-quality evidence on the statistically significant efficacy of morphine in 30% and 50% pain reduction. There is a lack of evidence on the efficacy of tramadol for other neuropathic pain conditions ( that is all other neuropthic pain conditions apart from PDN and PHN).

Overall Both tramadol and morphine seem to have some evidence on efficacy in different neuropathic pain conditions.

Topical treatments For post-herpetic neuralgia (PHN) The GDG agreed that there is high-quality evidence on the significant efficacy of topical capsaicin (8% patch) in 30% pain reduction and global improvement/impression of change, and low-quality evidence on 50% pain reduction for PHN. Moreover, this treatment is licensed for peripheral neuropathic pain (excluding PDN).

The GDG agreed that there was insufficient evidence on the efficacy of topical capsaicin (0.075% cream) and the low-quality evidence on topical lidocaine (5% patch) was on non-critical pain outcomes.

For other neuropathic pain conditions (other than PDN and PHN) The GDG agreed that there was insufficient evidence on the efficacy of

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 119 of 150

topical capsaicin (0.075% cream) and the low-quality evidence on topical lidocaine (5% patch) was on non-critical pain outcomes.

Overall The GDG agreed that there is high-quality evidence on the significant efficacy of topical capsaicin (8% patch), and limited low-quality evidence on non-critical outcome on pain for topical lidocaine (5% patch).

Trade-off between Desipramine clinical benefits Although there was some evidence for the efficacy of desipramine, it is no and harms longer in the BNF, and so should not be used in clinical practice.

Amitriptyline The GDG agreed that based on the evidence, amitriptyline should be recommended as one of the first-line and second-line treatments (as monotherapy or in combination with other first-line treatment) for PHN and for other neuropathic pain conditions (apart from PDN and PHN).

Although the GDG agreed with the role of amitriptyline, they were also concerned that many people who achieve satisfactory pain reduction with amitriptyline as first-line or second-line treatment would not be able to tolerate its adverse effects. The GDG reached a consensus that in these cases other TCAs, namely nortriptyline and imipramine, should be recommended as alternatives to amitriptyline, because there is evidence on imipramine in global improvement/impression of change, and nortriptyline was shown to be equally as effective as gabapentin in a head-to-head comparison study (see evidence statements 3.3.3.40 and 3.3.3.41). Both are relatively low-cost drugs, and for this patient population they are potentially cost effective, provided that they do not cause other adverse effects that would reduce the potential gain in quality of life obtained by switching from amitriptyline.

Gabapentin and pregabalin The GDG agreed that, based on the evidence, gabapentin and pregabalin should be recommended as one of the first-line and second-line treatments (as monotherapy or in combination with other first-line treatments) for PHN and other neuropathic pain conditions (apart from PDN and PHN). Both pregabalin and gabapentin have high affinity for the alpha-2-delta subunit of the voltage-dependent calcium channel in the central nervous system. Therefore if a person has unsatisfactory pain reduction with one drug, they are highly unlikely to have satisfactory pain

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 120 of 150

reduction with the other. Additionally, the cost-effectiveness evidence (see section below on economic considerations) demonstrated that gabapentin was more cost-effective than pregabalin for PHN. The GDG agreed that gabapentin should be a first-line treatment and a second-line treatment (as monotherapy or combination therapy with amitriptyline [or nortriptyline or imipramine]) for PHN and other neuropathic pain conditions (apart from PDN and PHN).

Because gabapentin and pregabalin have similar efficacy, the GDG also agreed that pregabalin should be an alternative if gabapentin is effective but the person cannot tolerate the adverse effects or has difficulty adhering to the dosage schedule.

Tramadol and morphine: The GDG agreed that the evidence on morphine and tramadol was limited. Because the GDG was concerned about the risk of long-term dependence, the severe adverse effects and the potential fatality of overdose with morphine, the GDG felt that morphine should not be initiated without an assessment by a specialist pain service or a condition- specific service.

However, the GDG also agreed that if patients did not have satisfactory pain reduction after first- and second-line treatment, they should be referred to a specialist pain service and/or a condition-specific service. Although evidence on tramadol was limited, the GDG came to the consensus that recommending tramadol is valid and appropriate as third- line treatment (as rescue analgesics) for PHN, PDN and other neuropathic pain conditions in non-specialist settings, either as monotherapy or as combination therapy with second-line treatment, because this drug is already commonly used in non-specialist settings. This will ensure continuity of treatment while a person is waiting for referral to a specialist pain service and/or a condition-specific service.

Topical capsaicin (8% patch) and topical lidocaine (5% patch): Although there is high-quality evidence on topical capsaicin (8% patch) for PHN and it is licensed for peripheral neuropathic pain (except for PDN), the GDG agreed that the application of topical capsaicin (8% patch) would need specialised training and would need the presence of a trained healthcare professional for 2 hours with the patient. Hence, currently topical capsaicin (8% patch) is not commonly used in non-specialist

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 121 of 150

settings.

The GDG agreed that topical capsaicin (8% patch) should not be initiated without an assessment by a specialist pain service or a condition-specific service. However, the GDG acknowledged that in the near future, there may be more healthcare professionals trained in using this treatment in non-specialist settings, and therefore the recommendation on topical capsaicin (8% patch) should be re-assessed during the routine 3-year review of this guideline.

The GDG agreed that there is only low-quality evidence on topical lidocaine (5% patch), and it was for non-critical pain outcomes. However, based on the clinical experience of members, the GDG acknowledged that a subgroup of people with ‘localised neuropathic pain’ who are unable to take oral medication because of medical conditions and/or disability may benefit from topical lidocaine. In view of the lack of moderate or high- quality evidence, the GDG felt that they could not recommend the use of topical lidocaine as first-line or second-line treatment. However, topical lidocaine may play a role as a rescue analgesic (while waiting for a referral to a specialist pain service) in a very small subgroup of people with localised pain who are unable to take oral medication because of medical conditions and/or disability.

Economic The GDG considered the results of the PHN and PDN analyses together considerations and noted that in both cases gabapentin was a cost-effective option. For PHN, gabapentin was the most cost effective option with the highest net monetary benefit and also the highest probability of being cost effective. Therefore, the GDG felt confident in recommending gabapentin as first- line treatment. It noted that, in the PHN analysis, pregabalin was the second most cost-effective option and that if gabapentin is effective, but the person cannot tolerate the adverse events, then pregabalin is likely to be the most cost-effective option for them. For similar reasons as for PDN, given the low acquisition cost of amitriptyline and similar efficacy to gabapentin and pregabalin, from the clinical review and GDG opinion amitriptyline would be a cost-effective alternative to gabapentin.

Other The GDG agreed that if first-line treatment did not result in satisfactory considerations pain reduction, a drug from another therapeutic class should be recommended as second-line treatment, either as monotherapy or as combination therapy with first-line treatment, instead of trying another drug

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 122 of 150

from the same therapeutic class.

Because amitriptyline is not licensed for neuropathic pain, the GDG came to the consensus that its initial dosage and titration should be lower than is recommended in the ‘British National Formulary’ (BNF).

Other treatments The GDG also came to the consensus that, to ensure continuity of care, pharmacological treatments other than those recommended in this guideline that are started in a specialist pain service or a condition-specific service may continue to be prescribed in non-specialist settings, with a multidisciplinary care plan, local shared care agreements and careful management of adverse effects.

Carbamazepine for the treatment of trigeminal neuralgia The GDG recognised that the evidence on carbamazepine for the treatment of neuropathic pain overall is very limited and dated. Therefore the GDG agreed that carbamazepine should not be recommended for use across all neuropathic pain conditions. Only one study on carbamazepine for treating trigeminal neuralgia met the inclusion and exclusion criteria of this guideline. However, the GDG acknowledged that carbamazepine (within its licensed indication) has been the routine treatment for trigeminal neuralgia in clinical practice since the 1960s. Anecdotal evidence from clinical experience also showed that carbamazepine may be effective for treating this condition. Because trigeminal neuralgia is an extremely painful condition, and there is no good-quality evidence on which to base specific recommendations for treating it, the GDG agreed that carbamazepine may have a specific role in treating trigeminal neuralgia, and expected that current practice will continue. Consequently, the GDG came to the consensus that a research recommendation should be made to further investigate the efficacy of carbamazepine for treating trigeminal neuralgia (see section 3.4.6).

1876

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 123 of 150

1877 3.4.6 Recommendations and research recommendations

1878 Recommendations

First-line treatment 1.1.10 Offer oral amitriptyline* or gabapentin as first-line treatment (see recommendation 1.1.13 for people with painful diabetic neuropathy).

1.1.11 For people with painful diabetic neuropathy, offer oral duloxetine as first-line treatment. If duloxetine is contraindicated, offer oral amitriptyline*.

1.1.12 Based on both the early and regular clinical reviews:

If improvement is satisfactory, continue the treatment; consider gradually reducing the dose over time if improvement is sustained. If amitriptyline* results in satisfactory pain reduction as first-line treatment but the person cannot tolerate the adverse effects, consider oral imipramine* or nortriptyline* as an alternative. If gabapentin results in satisfactory pain reduction as first-line treatment but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects, consider oral pregabalin as an alternative.

Second-line treatment 1.1.13 If satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person (see recommendation 1.1.16 for people with painful diabetic neuropathy):

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 124 of 150

If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with oral gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with gabapentin (or pregabalin) switch to or combine with oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects).

1.1.14 For people with painful diabetic neuropathy, if satisfactory pain reduction is not achieved with first-line treatment at the maximum tolerated dose, offer treatment with another drug; instead of or in combination with the original drug, after informed discussion with the person:

If first-line treatment was with duloxetine, switch to oral amitriptyline* (or imipramine* or nortriptyline* as an alternative if amitriptyline is effective but the person cannot tolerate the adverse effects) or switch to or combine with oral gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects). If first-line treatment was with amitriptyline* (or imipramine* or nortriptyline*), switch to or combine with gabapentin (or pregabalin as an alternative if gabapentin is effective but the person has difficulty adhering to the dosage schedule or cannot tolerate the adverse effects).

Third-line treatment 1.1.15 If satisfactory pain reduction is not achieved with second-line

* In these recommendations, drug names are marked with an asterisk if they do not have UK marketing authorisation for the indication in question at the time of publication (September 2011). Informed consent should be obtained and documented. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 125 of 150

treatment:

refer the person to a specialist pain service and/or a condition- specific service12 and while waiting for referral: consider oral tramadol as third-line treatment instead of or in combination13 with the second-line treatment consider a topical lidocaine patch for treatment of localised pain for people who are unable to take oral medication because of medical conditions and/or disability.

Other treatments 1.1.16 Do not start treatment with a topical capsaicin 8% patch or with opioids (such as morphine or oxycodone) other than tramadol without an assessment by a specialist pain service or a condition- specific service12.

1.1.17 Pharmacological treatments other than those recommended in this guideline that are started by a specialist pain service or a condition- specific service12 may continue to be prescribed in non-specialist settings, with a multidisciplinary care plan, local shared care agreements and careful management of adverse effects.

1879

1880 Research recommendations 1881 See appendix B for full details of research recommendations.

12 A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. 13 The combination of tramadol with amitriptyline, nortriptyline, imipramine or duloxetine is associated with a low risk of serotonin syndrome (the features of which include confusion, delirium, shivering, sweating, changes in blood pressure and myoclonus). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 126 of 150

1882 3.4.7 Evidence to recommendations (key principles of care)

1883 After the assessment and discussion of the evidence of the efficacy of 1884 different pharmacological treatments for neuropathic pain conditions (including 1885 PDN and PHN) and recommendations for treatment were derived, the GDG 1886 felt that patient’s care (other than the prescription of drugs) is also very 1887 important and that this should be further discussed in order to derive 1888 recommendations for good principles of care based on informal consensus. 1889 No evidence was considered within this section and therefore there were no 1890 evidence statements. The recommendations were based on the expertise and 1891 experience of the GDG.

Relative value of The GDG agreed that apart from the critical outcomes on pain and adverse different effects of pharmacological treatments (as stated in section 3.2.5 and 3.4.5), outcomes other elements of care such as patient’s experience, patient’s information needs, patient’s preference and different lifestyle factors are also important to be considered in a patient’s care pathway.

Quality of Not applicable evidence Trade-off between The GDG acknowledged that the low-quality evidence on adverse effects clinical benefits for both antidepressants and anti-epileptics was restricted by which and harms particular adverse effects were collected in the trials and how data on events were collected. Based on the knowledge and experience of GDG members in clinical practice, the evidence did not fully reflect a complete picture of the adverse effects that people would experience in real life. Issues such as the person’s vulnerability to specific adverse effects because of comorbidities, contraindications and safety considerations, current medication usage, mental health, lifestyle factors, daily activities and participation, patient preference and patients’ information needs should all be taken into consideration when selecting pharmacological treatments. The GDG further discussed that extra caution is needed when switching or combining drugs.

The GDG agreed that clear statements on drug dosage and titration in the recommendations are crucial for non-specialist settings, to emphasise the importance of titration to achieve maximum benefit. The GDG also agreed that the adverse effects of the recommended treatments, as well as the special warnings and precautions for its use as specified in the SPC (based

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 127 of 150

on advice from the Medicines and Healthcare Products Regulatory Agency [MHRA]), should be discussed with the person and weighed against the benefit provided.

Economic Not applicable considerations Other The GDG stressed that both early and regular clinical reviews are important considerations to assess the effectiveness of the treatment and to monitor drug titration, tolerability, adverse effects and the need to continue treatment (including the possibility of gradually reducing the dose if sustained improvement is observed). The GDG acknowledged that patient diaries may be a useful tool for recording progress and informing the clinical reviews. The principle of carrying out regular clinical reviews should apply to all treatments throughout the care pathway to ensure that people receive appropriate care.

Because referral to specialist pain services is not an exit from non-specialist care, but the start of a collaborative, ongoing approach to management, the GDG felt that the gateway for referrals to specialist pain services, as well as other condition-specific services, should not be at the end of the care pathway. Clinicians or healthcare professionals in non-specialist settings should consider making referrals at any stage of the care pathway, including at initial presentation and at the regular clinical reviews, if the person has severe pain or there are changes in, or deterioration of, the person’s pain, health condition, and/or daily activities and participation.

To ensure continuity of care, the GDG also came to a consensus that existing treatments should be continued for people whose neuropathic pain was already effectively managed before the publication of this guideline.

1892

1893 3.4.8 Recommendations and research recommendations

1894 Recommendations 1.1.1 Consider referring the person to a specialist pain service and/or a condition-specific service14 at any stage, including at initial

14 A condition-specific service is a specialist service that provides treatment for the underlying health condition that is causing neuropathic pain. Examples include neurology, diabetology and oncology services. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 128 of 150

presentation and at the regular clinical reviews (see recommendation 1.1.9), if:

they have severe pain or their pain significantly limits their daily activities and participation15 or their underlying health condition has deteriorated.

1.1.2 Continue existing treatments for people whose neuropathic pain is already effectively managed16.

1.1.3 Address the person’s concerns and expectations when agreeing which treatments to use by discussing:

the benefits and possible adverse effects of each pharmacological treatment why a particular pharmacological treatment is being offered coping strategies for pain and for possible adverse effects of treatment that non-pharmacological treatments are also available in non- specialist settings and/or through referral to specialist services (for example, surgical treatments and psychological therapies).

1.1.4 When selecting pharmacological treatments, take into account:

the person’s vulnerability to specific adverse effects because of comorbidities safety considerations and contraindications as detailed in the SPC

15 The World Health Organization ICF (International Classification of Functioning, Disability and Health) (2001) defines participation as ‘A person’s involvement in a life situation.’ It includes the following domains: learning and applying knowledge, general tasks and demands, mobility, self-care, domestic life, interpersonal interactions and relationships, major life areas, community, and social and civil life. 16 Note that there is currently no good-quality evidence on which to base specific recommendations for treating trigeminal neuralgia. The GDG expected that current routine practice will continue until new evidence is available (see also section 3.1). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 129 of 150

patient preference lifestyle factors (such as occupation) any mental health problems (such as depression and/or anxiety17) any other medication the person is taking.

1.1.5 Explain both the importance of dosage titration and the titration process, providing written information if possible.

1.1.6 When withdrawing or switching treatment, taper the withdrawal regimen to take account of dosage and any discontinuation symptoms.

1.1.7 When introducing a new treatment, consider overlap with the old treatments to avoid deterioration in pain control.

1.1.8 After starting or changing a treatment, perform an early clinical review of dosage titration, tolerability and adverse effects to assess the suitability of the chosen treatment.

1.1.9 Perform regular clinical reviews to assess and monitor the effectiveness of the chosen treatment. Each review should include assessment of:

pain reduction adverse effects daily activities and participation18 (such as ability to work and drive) mood (in particular, whether the person may have depression

17 Refer if necessary to ‘Anxiety’ (NICE clinical guideline 113), ‘Depression’ (NICE clinical guideline 90) and/or ‘Depression in adults with a chronic physical health problem’ (NICE clinical guideline 91) (available at www.nice.org.uk). 18 The World Health Organization ICF (International Classification of Functioning, Disability and Health) (2001) defines participation as ‘A person’s involvement in a life situation.’ It includes the following domains: learning and applying knowledge, general tasks and demands, mobility, self-care, domestic life, interpersonal interactions and relationships, major life areas, community, and social and civil life. Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 130 of 150

and/or anxiety19) quality of sleep overall improvement as reported by the person.

1895

1896 4 Notes on the scope of the guideline

1897 NICE guidelines are developed in accordance with a scope that defines what 1898 the guideline will and will not cover. The scope of this guideline is given in 1899 appendix C.

1900 5 Implementation

1901 NICE has developed tools to help organisations implement this guidance (see 1902 www.nice.org.uk/guidance/CG[xxx])’. Note: these details will apply when the 1903 guideline is published.

1904 6 Other versions of this guideline

1905 6.1 Quick reference guide

1906 A quick reference guide for healthcare professionals is available from 1907 www.nice.org.uk/guidance/CG96/QuickRefGuide

1908 For printed copies, phone NICE publications on 0845 003 7783 or email 1909 [email protected] (quote reference number N[xxxx]). Note: these 1910 details will apply when the guideline is published.

1911 6.2 NICE pathway

1912 The recommendations from this guideline have been incorporated into a NICE 1913 pathway, which is available from http://pathways.nice.org.uk/pathways/[xxx]

19 Refer if necessary to ‘Anxiety’ (NICE clinical guideline 113), ‘Depression’ (NICE clinical guideline 90) and/or ‘Depression in adults with a chronic physical health problem’ (NICE clinical guideline 91) (available at www.nice.org.uk). Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 131 of 150

1914 Note: these details will apply when the guideline is published.

1915 6.3 ‘Understanding NICE guidance’

1916 A summary for patients and carers (‘Understanding NICE guidance’) is 1917 available from www.nice.org.uk/guidance/CG[xxx]/PublicInfo

1918 For printed copies, phone NICE publications on 0845 003 7783 or email 1919 [email protected] (quote reference number N[xxxx]). Note: these 1920 details will apply when the guideline is published.

1921 We encourage NHS and voluntary sector organisations to use text from this 1922 booklet in their own information about neuropathic pain.

1923 7 Related NICE guidance

1924 Published 1925 Anxiety. NICE clinical guideline 113 (2011). Available from 1926 www.nice.org.uk/guidance/CG113 1927 Depression in adults with a chronic physical health problem. NICE clinical 1928 guideline 91 (2009). Available from www.nice.org.uk/guidance/CG91 1929 Depression. NICE clinical guideline 90 (2009). Available from 1930 www.nice.org.uk/guidance/CG90 1931 Type 2 diabetes. NICE clinical guideline 87 (2009). Available from 1932 www.nice.org.uk/guidance/CG87 1933 Medicines adherence. NICE clinical guideline 76 (2009). Available from 1934 www.nice.org.uk/guidance/CG76 1935 Spinal cord stimulation for chronic pain of neuropathic or ischaemic origin. 1936 NICE technology appraisal guidance 159 (2008). Available from 1937 www.nice.org.uk/guidance/TA159 1938 Type 1 diabetes. NICE clinical guideline 15 (2004; amended 2009). 1939 Available from www.nice.org.uk/guidance/CG15

1940 8 Updating the guideline

1941 NICE clinical guidelines are updated so that recommendations take into 1942 account important new information. New evidence is checked 3 years after Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 132 of 150

1943 publication, and healthcare professionals and patients are asked for their 1944 views; we use this information to decide whether all or part of a guideline 1945 needs updating. If important new evidence is published at other times, we 1946 may decide to do a more rapid update of some recommendations. Please see 1947 our website for information about updating the guideline.

1948 9 References

1949 Agrawal RP, Goswami J, Jain S et al. (2009) Management of diabetic 1950 neuropathy by sodium valproate and glyceryl trinitrate spray: a prospective 1951 double-blind randomized placebo-controlled study. Diabetes Research and 1952 Clinical Practice 83: 371–8

1953 Arbaiza D, Vidal O (2007) Tramadol in the treatment of neuropathic cancer 1954 pain: a double-blind, placebo-controlled study. Clinical Drug Investigation 27: 1955 75–83

1956 Arezzo JC, Rosenstock J, Lamoreaux L et al. (2008) Efficacy and safety of 1957 pregabalin 600 mg/d for treating painful diabetic peripheral neuropathy: a 1958 double-blind placebo-controlled trial. BMC Neurology 8: 33

1959 Backonja M, Beydoun A, Edwards KR et al. (1998) Gabapentin for the 1960 symptomatic treatment of painful neuropathy in patients with diabetes mellitus. 1961 A randomized controlled trial. Journal of the American Medical Association 1962 280: 1831–6

1963 Baron R, Mayoral V, Leijon G et al. (2009) Efficacy and safety of 5% lidocaine 1964 (lignocaine) medicated plaster in comparison with pregabalin in patients with 1965 postherpetic neuralgia and diabetic polyneuropathy: interim analysis from an 1966 open-label, two-stage adaptive, randomized, controlled trial. Clinical Drug 1967 Investigation 29: 231–41

1968 Barton GR, Briggs AH, Fenwick EA (2008) Optimal cost-effectiveness 1969 decisions: the role of the cost-effectiveness acceptability curve (CEAC), the 1970 cost-effectiveness acceptability frontier (CEAF), and the expected value of 1971 perfection information (EVPI). Value Health 11: 886–97

1972 Beniczky S, Tajti J, Timea VE et al. (2005) Evidence-based pharmacological 1973 treatment of neuropathic pain syndromes. Journal of Neural Transmission 1974 112: 735–49

1975 Bennett GJ (1997) Neuropathic pain: an overview. In: Borsook D, editor. 1976 Molecular Biology of Pain. Seattle: IASP Press; p109–13

1977 Bernstein JE, Korman NJ, Bickers DR et al. (1989) Topical capsaicin 1978 treatment of chronic postherpetic neuralgia. Journal of the American Academy 1979 of Dermatology 21: 265–70.

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 133 of 150

1980 Beydoun A, Shaibani A, Hopwood M et al. (2006) Oxcarbazepine in painful 1981 diabetic neuropathy: results of a dose-ranging study. Acta Neurologica 1982 Scandinavica 113: 395–404

1983 Biesbroeck R, Bril V, Hollander P et al. (1995) A double-blind comparison of 1984 topical capsaicin and oral amitriptyline in painful diabetic neuropathy. 1985 Advances in Therapy 12: 111–20

1986 Bone M, Critchley P, Buggy DJ (2002) Gabapentin in postamputation 1987 phantom limb pain: a randomized, double-blind, placebo-controlled, cross- 1988 over study. Regional Anesthesia and Pain Medicine 27: 481–6

1989 Boureau F, Legallicier P, Kabir-Ahmadi M (2003) Tramadol in post-herpetic 1990 neuralgia: a randomized, double-blind, placebo-controlled trial. Pain 104: 323– 1991 31

1992 Bowsher D, Rigge M, Sopp L (1991) Prevalence of chronic pain in the British 1993 population: a telephone survey of 1037 households. The Pain Clinic 4: 223–30

1994 Bowsher D (1997) The effects of pre-emptive treatment of postherpetic 1995 neuralgia with amitriptyline: a randomized, double-blind, placebo-controlled 1996 trial. Journal of Pain & Symptom Management 13: 327–31

1997 Breuer B, Pappagallo M, Knotkova H et al. (2007) A randomized, double- 1998 blind, placebo-controlled, two-period, crossover, pilot trial of lamotrigine in 1999 patients with central pain due to multiple sclerosis. Clinical Therapeutics 29: 2000 2022–30

2001 Briggs A, Sculpher M (1998) An introduction to Markov modelling for 2002 economic evaluation. Pharmacoeconomics 13: 397-–409

2003 Briggs AH, Sculpher M, Claxton K (2006) Decision modelling for health 2004 economic evaluation. Oxford: Oxford University Press

2005 Cardenas DD, Warms CA, Turner JA et al. (2002) Efficacy of amitriptyline for 2006 relief of pain in spinal cord injury: results of a randomized controlled trial. Pain 2007 96: 365–73

2008 Chandra K, Shafiq N, Pandhi P et al. (2006) Gabapentin versus nortriptyline in 2009 post-herpetic neuralgia patients: a randomized, double-blind clinical trial--the 2010 GONIP Trial. International Journal of Clinical Pharmacology and Therapeutics 2011 44: 358–63

2012 Cheville AL, Sloan JA, Northfelt DW et al. (2009) Use of a lidocaine patch in 2013 the management of postsurgical neuropathic pain in patients with cancer: a 2014 phase III double-blind crossover study (N01CB). Supportive Care in Cancer 2015 17: 451–60

2016 Claxton K (2008) Exploring uncertainty in cost-effectiveness analysis. 2017 Pharmacoeconomics 26: 781–98

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 134 of 150

2018 Coyle D, Oakley J (2008) Estimating the expected value of partial perfect 2019 information: a review of methods. The European Journal of Health Economics 2020 9: 251–9

2021 Dallocchio C, Buffa C, Mazzarello P et al. (2000) Gabapentin vs. amitriptyline 2022 in painful diabetic neuropathy: an open-label pilot study. Journal of Pain & 2023 Symptom Management 20: 280–5

2024 Dieleman JP, Kerklaan J, Huygen FJ et al. (2008) Incidence rates and 2025 treatment of neuropathic pain conditions in the general population. Pain 31: 2026 137: 681–8

2027 Dogra S, Beydoun S, Mazzola J et al. (2005) Oxcarbazepine in painful 2028 diabetic neuropathy: A randomized, placebo-controlled study. European 2029 Journal of Pain 9: 543–54

2030 Donofrio P, Walker F, Hunt V et al. (1991) Treatment of painful diabetic 2031 neuropathy with topical capsaicin: A multicenter, double-blind, vehicle- 2032 controlled study. Archives of Internal Medicine 151: 2225–2229

2033 Drummond MF, Sculpher MJ, Torrance GW et al. (2005) Methods for the 2034 economic evaluation of health care programmes. Oxford: Oxford University 2035 Press

2036 Dworkin RH, Corbin AE, Young JP Jr. et al. (2003) Pregabalin for the 2037 treatment of postherpetic neuralgia: a randomized, placebo-controlled trial. 2038 Neurology 60: 1274–83

2039 Dworkin RH, Turk DC, Farrar JT et al. (2005) Core outcome measures for 2040 chronic pain clinical trials: IMMPACT recommendations. Pain 113: 9–19

2041 Dworkin RH, O'Connor AB, Backonja M et al. (2007) Pharmacologic 2042 management of neuropathic pain: evidence-based recommendations. Pain 2043 132: 237–51

2044 Dworkin RH, Turk DC, Wyrwich KW et al. (2008) Interpreting the clinical 2045 importance of treatment outcomes in chronic pain clinical trials: IMMPACT 2046 recommendations. Journal of Pain 9:105–21.

2047 Eisenberg E, Lurie Y, Braker C et al. (2001) Lamotrigine reduces painful 2048 diabetic neuropathy: a randomized, controlled study. Neurology 57: 505–9

2049 Estanislao L, Carter K, McArthur J et al. (2004) A randomized controlled trial 2050 of 5% lidocaine gel for HIV-associated distal symmetric polyneuropathy. 2051 Journal of Acquired Immune Deficiency Syndromes 37: 1584–6

2052 Finnerup NB, Sindrup SH, Bach FW et al. (2002) Lamotrigine in spinal cord 2053 injury pain: a randomized controlled trial. Pain 96: 375–83.

2054 Fox-Rushby JA, GL Griffith, JR Ross et al. (2010) The clinical and cost- 2055 effectiveness of different treatment pathways for neuropathic pain [NP]. NIHR

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 135 of 150

2056 Health Technology Assessment (HTA) programme, ref. 05/30/03. In press. 2057 Available from www.hta.ac.uk/1527

2058 Freynhagen R, Strojek K, Griesing T et al. (2005) Efficacy of pregabalin in 2059 neuropathic pain evaluated in a 12-week, randomised, double-blind, 2060 multicentre, placebo-controlled trial of flexible- and fixed-dose regimens. Pain 2061 115: 254–63

2062 Galer BS, Jensen MP, Ma T et al. (2002) The lidocaine patch 5% effectively 2063 treats all neuropathic pain qualities: results of a randomized, double-blind, 2064 vehicle-controlled, 3-week efficacy study with use of the neuropathic pain 2065 scale. Clinical Journal of Pain 18: 297–301

2066 Gatti A, Sabato AF, Occhioni R et al. (2009) Controlled-release oxycodone 2067 and pregabalin in the treatment of neuropathic pain: Results of a multicenter 2068 Italian study. European Neurology 61: 129–37

2069 Gilron I, Bailey JM, Tu D et al. (2009) Nortriptyline and gabapentin, alone and 2070 in combination for neuropathic pain: a double-blind, randomised controlled 2071 crossover trial. Lancet 374: 1252–61

2072 Gimbel JS, Richards P, Portenoy RK (2003) Controlled-release oxycodone for 2073 pain in diabetic neuropathy: a randomized controlled trial. Neurology 60: 927– 2074 34

2075 Goldstein DJ, Lu Y, Detke MJ et al. (2005) Duloxetine vs. placebo in patients 2076 with painful diabetic neuropathy. Pain 116: 109–18

2077 Gordh TE, Stubhaug A, Jensen TS et al. (2008) Gabapentin in traumatic 2078 nerve injury pain: a randomized, double-blind, placebo-controlled, cross-over, 2079 multi-center study. Pain 138: 255–66

2080 Graff-Radford SB, Shaw LR, Naliboff BN (2000) Amitriptyline and 2081 fluphenazine in the treatment of postherpetic neuralgia. Clinical Journal of 2082 Pain 16: 188–92

2083 Grosskopf J, Mazzola J, Wan Y et al. (2006) A randomized, placebo- 2084 controlled study of oxcarbazepine in painful diabetic neuropathy. Acta 2085 Neurologica Scandinavica 114: 177–80

2086 Hahn K, Arendt G, Braun JS et al. (2004) A placebo-controlled trial of 2087 gabapentin for painful HIV-associated sensory neuropathies. Journal of 2088 Neurology 251: 1260–6

2089 Hanna M, O'Brien C, Wilson MC (2008) Prolonged-release oxycodone 2090 enhances the effects of existing gabapentin therapy in painful diabetic 2091 neuropathy patients. European Journal of Pain 12: 804–13

2092 Harati Y, Gooch C, Swenson M et al. (1998) Double-blind randomized trial of 2093 tramadol for the treatment of the pain of diabetic neuropathy. Neurology 50: 2094 1842–6

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 136 of 150

2095 Ho KY, Huh BK, White WD et al. (2008) Topical amitriptyline versus lidocaine 2096 in the treatment of neuropathic pain. The Clinical Journal of Pain 24: 51–5

2097 Hoch JS, Briggs AH, Willan AR (2002) Something old, something new, 2098 something borrowed, something blue: a framework for the marriage of health 2099 econometrics and cost-effectiveness analysis. Health Economics 11: 415–30

2100 Huse E, Larbig W, Flor H et al. (2001) The effect of opioids on phantom limb 2101 pain and cortical reorganization. Pain 90: 47–55

2102 International Association for the Study of Pain (2007) IASP Pain terminology 2103 [online]. Available from www.iasp- 2104 pain.org/AM/Template.cfm?Section=Home&Template=/CM/HTMLDisplay.cfm 2105 &ContentID=3058#Neuropathic [accessed 26 August 2009]

2106 Jensen TS, Backonja MM, Hernandez Jimenez S et al. (2006) New 2107 perspectives on the management of diabetic peripheral neuropathic pain. 2108 Diabetes & Vascular Disease Research 3: 108–19

2109 Jung BF, Johnson RW, Griffin DR et al. (2004) Risk factors for postherpetic 2110 neuralgia in patients with herpes zoster. Neurology 62: 1545–51

2111 Kalso E, Tasmuth T, Neuvonen PJ (1996) Amitriptyline effectively relieves 2112 neuropathic pain following treatment of breast cancer. Pain 64: 293–302

2113 Kautio AL, Haanpaa M, Saarto T et al. (2008) Amitriptyline in the treatment of 2114 chemotherapy-induced neuropathic symptoms. Journal of Pain and Symptom 2115 Management 35: 31–9

2116 Kehlet H, Jensen TS, Woolf CJ (2006) Persistent postsurgical pain: risk 2117 factors and prevention. Lancet 367: 1618–25

2118 Khoromi S, Patsalides A, Parada S et al. (2005) Topiramate in chronic lumbar 2119 radicular pain. The Journal of Pain: Official Journal of the American Pain 2120 Society 6: 829–36

2121 Khoromi S, Cui L, Nackers L et al. (2007) Morphine, nortriptyline and their 2122 combination vs. placebo in patients with chronic lumbar root pain. Pain 130: 2123 66–75

2124 Kieburtz K, Simpson D, Yiannoutsos C et al. (1998) A randomized trial of 2125 amitriptyline and mexiletine for painful neuropathy in HIV infection. Neurology 2126 51: 1682–8

2127 Kishore-Kumar R, Max MB, Schafer SC et al. (1990) Desipramine relieves 2128 postherpetic neuralgia. Clinical Pharmacology & Therapeutics 47: 305–12

2129 Kochar DK, Jain N, Agarwal RP et al. (2002) Sodium valproate in the 2130 management of painful neuropathy in type 2 diabetes - a randomized placebo 2131 controlled study. Acta Neurologica Scandinavica 106: 248–52

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 137 of 150

2132 Kochar DK, Rawat N, Agrawal RP et al. (2004) Sodium valproate for painful 2133 diabetic neuropathy: A randomized double-blind placebo-controlled study. 2134 QJM: An International Journal of Medicine 97: 33–8

2135 Leijon G, Boivie J (1989) Central post-stroke pain--a controlled trial of 2136 amitriptyline and carbamazepine. Pain 36: 27–36

2137 Lesser H, Sharma U, Lamoreaux L et al. (2004) Pregabalin relieves 2138 symptoms of painful diabetic neuropathy: a randomized controlled trial. 2139 Neurology 63: 2104–10

2140 Levendoglu F, Ogun CO, Ozerbil O et al. (2004) Gabapentin is a first line drug 2141 for the treatment of neuropathic pain in spinal cord injury. Spine 29: 743–51

2142 Low PA, Opfer-Gehrking TL, Dyck PJ et al. (1995) Double-blind, placebo- 2143 controlled study of the application of capsaicin cream in chronic distal painful 2144 polyneuropathy. Pain 62: 163–8

2145 Luria Y, Brecker C, Daoud D et al. (2000) Lamotrigine in the treatment of 2146 painful diabetic neuropathy: A randomized, placebo-controlled study. Progress 2147 in Pain Research and Management 16: 857–62

2148 Max MB, Schafer SC, Culnane M et al. (1988) Amitriptyline, but not 2149 lorazepam, relieves postherpetic neuralgia. Neurology 38: 1427–32

2150 Max MB, Kishore-Kumar R, Schafer SC et al. (1991) Efficacy of desipramine 2151 in painful diabetic neuropathy: a placebo-controlled trial. Pain 45: 3–9

2152 McCarberg B (2006) Pharmacotherapy for neuropathic pain: The old and the 2153 new. Advanced Studies in Medicine 6: 399–408

2154 McCleane G (1999) 200 mg daily of lamotrigine has no analgesic effect in 2155 neuropathic pain: a randomised, double-blind, placebo controlled trial. Pain 2156 83: 105–7

2157 McCleane G (2000) The analgesic efficacy of topical capsaicin is enhanced by 2158 glyceryl trinitrate in painful osteoarthritis: a randomized, double blind, placebo 2159 controlled study. European Journal of Pain 4: 355–60

2160 Meier T, Wasner G, Faust M et al. (2003) Efficacy of lidocaine patch 5% in the 2161 treatment of focal peripheral neuropathic pain syndromes: a randomized, 2162 double-blind, placebo-controlled study. Pain 106: 151–8

2163 Mikkelsen T, Werner MU, Lassen B et al. (2004) Pain and sensory 2164 dysfunction 6 to 12 months after inguinal herniotomy. Anesthesia Analgesia 2165 99: 146–51

2166 Morello CM, Leckband SG, Stoner CP et al. (1999) Randomized double-blind 2167 study comparing the efficacy of gabapentin with amitriptyline on diabetic 2168 peripheral neuropathy pain. Archives of Internal Medicine 159: 1931–7

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 138 of 150

2169 National Institute for Health and Clinical Excellence (2009) The guidelines 2170 manual. London: National Institute for Health and Clinical Excellence. 2171 Available from: www.nice.org.uk/GuidelinesManual

2172 Nicol CF (1969) A four year double-blind study of tegretol in facial pain. 2173 Headache 9: 54–7

2174 Nikolajsen L, Finnerup NB, Kramp S et al. (2006) A randomized study of the 2175 effects of gabapentin on postamputation pain. Anesthesiology 105: 1008–15

2176 Paice JA, Ferrans CE, Lashley FR et al. (2000) Topical capsaicin in the 2177 management of HIV-associated peripheral neuropathy. Journal of Pain and 2178 Symptom Management 19: 45–52

2179 Rao RD, Michalak JC, Sloan JA et al. (2007) Efficacy of gabapentin in the 2180 management of chemotherapy-induced peripheral neuropathy: a phase 3 2181 randomized, double-blind, placebo-controlled, crossover trial (N00C3). Cancer 2182 110: 2110–8

2183 Rao RD, Flynn PJ, Sloan JA et al. (2008) Efficacy of lamotrigine in the 2184 management of chemotherapy-induced peripheral neuropathy: a phase 3 2185 randomized, double-blind, placebo-controlled trial, N01C3. Cancer 112: 2802– 2186 8

2187 Raskin P, Donofrio PD, Rosenthal NR et al. (2004) Topiramate vs placebo in 2188 painful diabetic neuropathy: analgesic and metabolic effects. Neurology 63: 2189 865–73

2190 Raskin J, Pritchett Y, Chappell AS et al. (2005) Duloxetine in the treatment of 2191 diabetic peripheral neuropathic pain - results from three clinical trials. Poster 2192 presented at the 9th Congress of the European Federation of Neurological 2193 Societies; 17–20 September 2005, Athens, Greece

2194 Rice AS, Maton S (2001) Gabapentin in postherpetic neuralgia: A 2195 randomised, double blind, placebo controlled study. Pain 94: 215–24

2196 Richter RW, Portenoy R, Sharma U et al. (2005) Relief of painful diabetic 2197 peripheral neuropathy with pregabalin: a randomized, placebo-controlled trial. 2198 The Journal of Pain: Official Journal of the American Pain Society 6: 253–60

2199 Rintala DH, Holmes SA, Courtade D et al. (2007) Comparison of the 2200 effectiveness of amitriptyline and gabapentin on chronic neuropathic pain in 2201 persons with spinal cord injury. Archives of Physical Medicine and 2202 Rehabilitation 88: 1547–60 (erratum in Archives of Physical Medicine and 2203 Rehabilitation 89: 1206)

2204 Robinson LR, Czerniecki JM, Ehde DM et al. (2004) Trial of amitriptyline for 2205 relief of pain in amputees: results of a randomized controlled study. Archives 2206 of Physical Medicine and Rehabilitation 85: 1–6

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 139 of 150

2207 Rosenstock J, Tuchman M, Lamoreaux L et al. (2004) Pregabalin for the 2208 treatment of painful diabetic peripheral neuropathy: a double-blind, placebo- 2209 controlled trial. Pain 110: 628–38

2210 Rowbotham M, Harden N, Stacey B et al. (1998) Gabapentin for the treatment 2211 of postherpetic neuralgia: a randomized controlled trial. JAMA: the Journal of 2212 the American Medical Association 280: 1837–42

2213 Rowbotham MC, Goli V, Kunz NR et al. (2004) Venlafaxine extended release 2214 in the treatment of painful diabetic neuropathy: a double-blind, placebo- 2215 controlled study. Pain 110: 697–706 (erratum in Pain 113: 248)

2216 Sabatowski R, Galvez R, Cherry DA et al. (2004) Pregabalin reduces pain and 2217 improves sleep and mood disturbances in patients with post-herpetic 2218 neuralgia: results of a randomised, placebo-controlled clinical trial. Pain 109: 2219 26–35

2220 Schmader KE (2002) Epidemiology and impact on quality of life of 2221 postherpetic neuralgia and painful diabetic neuropathy. The Clinical Journal of 2222 Pain 18: 350–4

2223 Scheffler NM, Sheitel PL, Lipton MN (1991) Treatment of painful diabetic 2224 neuropathy with capsaicin 0.075%. Journal of the American Podiatric Medical 2225 Association 81: 288–93

2226 Serpell MG Neuropathic pain study group (2002) Gabapentin in neuropathic 2227 pain syndromes: a randomised, double-blind, placebo-controlled trial. Pain 99: 2228 557–66

2229 Shipton E (2008) Post-surgical neuropathic pain. ANZ Journal of Surgery 78: 2230 548–55

2231 Siddall PJ, Cousins MJ, Otte A et al. (2006) Pregabalin in central neuropathic 2232 pain associated with spinal cord injury: a placebo-controlled trial. Neurology 2233 67: 1792–800

2234 Simpson DA (2001) Gabapentin and venlafaxine for the treatment of painful 2235 diabetic neuropathy. Journal of Clinical Neuromuscular Disease 3: 53–62

2236 Simpson DM, Olney R, McArthur JC et al. (2000) A placebo-controlled trial of 2237 lamotrigine for painful HIV-associated neuropathy. Neurology 54: 2115–9

2238 Simpson DM, McArthur JC, Olney R et al. (2003) Lamotrigine for HIV- 2239 associated painful sensory neuropathies: a placebo-controlled trial. Neurology 2240 60: 1508–14

2241 Sindrup SH, Bach FW, Madsen C et al. (2003) Venlafaxine versus imipramine 2242 in painful polyneuropathy: a randomized, controlled trial. Neurology 60: 1284– 2243 9

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 140 of 150

2244 Smith DG, Ehde DM, Hanley MA et al. (2005) Efficacy of gabapentin in 2245 treating chronic phantom limb and residual limb pain. Journal of Rehabilitation 2246 Research & Development 42: 645–54

2247 Smith BH, Torrance N (2010) Neuropathic pain. In: Croft PR, editor. Chronic 2248 pain epidemiology: from aetiology to public health. Oxford: Oxford University 2249 Press, in press (ISBN 9780199235766)

2250 Stacey BR, Barrett JA, Whalen E et al. (2008) Pregabalin for postherpetic 2251 neuralgia: placebo-controlled trial of fixed and flexible dosing regimens on 2252 allodynia and time to onset of pain relief. Journal of Pain 9: 1006–17

2253 Tandan R, Lewis GA, Krusinski PB et al. (1992) Topical capsaicin in painful 2254 diabetic neuropathy. Controlled study with long-term follow-up. Diabetes Care 2255 15: 8–14

2256 Tasmuth T, Hartel B, Kalso E (2002) Venlafaxine in neuropathic pain following 2257 treatment of breast cancer. European Journal of Pain 6: 17–24

2258 Thienel U, Neto W, Schwabe SK et al. (2004) Topiramate in painful diabetic 2259 polyneuropathy: findings from three double-blind placebo-controlled trials. 2260 Acta Neurologica Scandinavica 110: 221–31

2261 Tölle T, Freynhagen R, Versavel M et al. (2008) Pregabalin for relief of 2262 neuropathic pain associated with diabetic neuropathy: A randomized, double- 2263 blind study. European Journal of Pain 12: 203–13

2264 van Seventer R, Sadosky A, Lucero M et al. (2006) A cross-sectional survey 2265 of health state impairment and treatment patterns in patients with postherpetic 2266 neuralgia. Age and Ageing 35: 132–7

2267 Vestergaard K, Andersen G, Gottrup H et al. (2001) Lamotrigine for central 2268 poststroke pain: a randomized controlled trial. Neurology 56: 184–90

2269 Vinik AI, Tuchman M, Safirstein B et al. (2007) Lamotrigine for treatment of 2270 pain associated with diabetic neuropathy: results of two randomized, double- 2271 blind, placebo-controlled studies. Pain 128: 169–79

2272 Vranken JH, Dijkgraaf MG, Kruis MR et al. (2008) Pregabalin in patients with 2273 central neuropathic pain: a randomized, double-blind, placebo-controlled trial 2274 of a flexible-dose regimen. Pain 136: 150–7

2275 Vrethem M, Boivie J, Arnqvist H et al. (1997) A comparison a amitriptyline and 2276 maprotiline in the treatment of painful polyneuropathy in diabetics and 2277 nondiabetics. Clinical Journal of Pain 13: 313–23

2278 Wailoo AJ, Sutton AJ, Cooper NJ et al. (2008) Cost-effectiveness and value of 2279 information analyses of neuraminidase inhibitors for the treatment of 2280 influenza. Value Health. 11: 160–71

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 141 of 150

2281 Watson CP, Evans RJ (1992) The postmastectomy pain syndrome and topical 2282 capsaicin: a randomized trial. Pain 51: 375–9

2283 Watson CP, Tyler KL, Bickers DR et al. (1993) A randomized vehicle- 2284 controlled trial of topical capsaicin in the treatment of postherpetic neuralgia. 2285 Clinical Therapeutics 15: 510–26

2286 Watson CP, Vernich L, Chipman M et al. (1998) Nortriptyline versus 2287 amitriptyline in postherpetic neuralgia: a randomized trial. Neurology 51: 2288 1166–71

2289 Wernicke JF, Pritchett YL, D'Souza DN et al. (2006) A randomized controlled 2290 trial of duloxetine in diabetic peripheral neuropathic pain. Neurology 67: 1411– 2291 20

2292 World Health Organization (2007) International Statistical Classification of 2293 Diseases and Related Health Problems (ICD), 10th revision. Available at: 2294 http://apps.who.int/classifications/apps/icd/icd10online/

2295 Wu CL, Agarwal S, Tella PK et al. (2008) Morphine versus mexiletine for 2296 treatment of postamputation pain: a randomized, placebo-controlled, 2297 crossover trial. Anesthesiology 109: 289–96

2298 Yucel A, Ozyalcin S, Koknel TG et al. (2005) The effect of venlafaxine on 2299 ongoing and experimentally induced pain in neuropathic pain patients: a 2300 double blind, placebo controlled study. European Journal of Pain 9: 407–16

2301 Ziegler D (2008) Painful diabetic neuropathy: treatment and future aspects. 2302 Diabetes/Metabolism Research and Reviews 24 (Suppl. 1): S52–7.

2303

2304

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 142 of 150

2305 10 Glossary and abbreviations

2306 Glossary

Absolute risk Measures the probability of an event or outcome occurring (e.g. an adverse reaction to the drug being tested) in the group of people under study. Studies that compare two or more groups of patients may report results in terms of the absolute risk reduction.

Absolute risk reduction (ARR) (risk difference) The ARR is the difference in the risk of an event occurring between two groups of patients in a study – for example if 6% of patients die after receiving a new experimental drug and 10% of patients die after having the old drug treatment then the ARR is 10% - 6% = 4%. Thus by using the new drug instead of the old drug 4% of patients can be prevented from dying. Here the ARR measures the risk reduction associated with a new treatment. See also absolute risk.

Absolute risk increase (risk difference) Same as ARR but with different direction of effect.

Bias Influences on a study that can lead to invalid conclusions about a treatment or intervention. Bias in research can make a treatment look better or worse than it really is. Bias can even make it look as if the treatment works when it actually doesn't. Bias can occur by chance or as a result of systematic errors in the design and execution of a study. Bias can occur at different stages in the research process, for example in the collection, analysis, interpretation, publication or review of research data. For examples, see selection bias, performance bias, information bias, confounding, publication bias.

Clinical effectiveness The extent to which a specific treatment or intervention, when used under usual or everyday conditions, has a beneficial effect on the course or outcome of disease compared with no treatment or other routine care. Clinical trials that assess effectiveness are sometimes called management trials. Clinical 'effectiveness' is not the same as efficacy.

Comorbidity Co-existence of a disease or diseases in the people being studied in addition to the health problem that is the subject of the study.

Confidence interval A way of expressing certainty about the findings from a study or group of studies, using statistical techniques. A confidence interval describes a range of possible effects (of a treatment or

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 143 of 150 intervention) that are consistent with the results of a study or group of studies. A wide confidence interval indicates a lack of certainty or precision about the true size of the clinical effect and is seen in studies with too few patients. If confidence intervals are narrow they indicate more precise estimates of effects and a larger sample of patients studied. It is usual to interpret a '95%' confidence interval as the range of effects within which we are 95% confident that the true effect lies.

Cost-effectiveness analysis An economic evaluation that compares alternative options for a specific patient group looking at a single effectiveness dimension measured in a non-monetary (natural) unit. It expresses the result in the form of an incremental (or average or marginal) cost-effectiveness ratio.

Economic evaluation A comparison of alternative courses of action in terms of both their costs and consequences. In health economic evaluations the consequences should include health outcomes.

Guideline Development Group A group of healthcare professionals, patients, carers and technical staff who develop the recommendations for a clinical guideline. The short clinical guidelines team or national collaborating centre responsible for developing the guideline recruits a guideline development group to work on the guideline. Staff from the short guidelines team or the national collaborating centre review the evidence and support the guideline development group. The group writes draft guidance, and then revises it after a consultation with organisations registered as stakeholders.

Generalisability The extent to which the results of a study hold true for a population of patients beyond those who participated in the research. See also external validity.

Heterogeneity Or lack of homogeneity. The term is used in meta-analyses and systematic reviews when the results or estimates of effects of treatment from separate studies seem to be very different - in terms of the size of treatment effects or even to the extent that some indicate beneficial and others suggest adverse treatment effects. Such results may occur as a result of differences between studies in terms of the patient populations, outcome measures, definition of variables or duration of follow-up.

Number needed to treat to benefit (NNTB) This measures the impact of a treatment or intervention. It states how many patients need to be treated with the treatment in question in order to prevent an event which would otherwise occur. For example, if the NNTB = 4, then four patients would have to be treated to prevent one bad outcome. The closer the NNTB is to 1, the better the treatment is. Analogous to the NNTB is the Number needed to treat to harm (NNTH), which is the number of patients that would need to receive a

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 144 of 150

treatment to cause one additional adverse event. For example if the NNTH = 4, then four patients would have to be treated for one bad outcome to occur.

Number needed to treat to harm (NNTH) See NNTB.

Quality-adjusted life year (QALY) A measure of health outcome which looks at both length of life and quality of life. QALYS are calculated by estimating the years of life remaining for a patient following a particular care pathway and weighting each year with a quality of life score (on a zero to one scale). One QALY is equal to one year of life in perfect health, or 2 years at 50% health, and so on.

Randomised controlled trial A study to test a specific drug or other treatment in which people are randomly assigned to two (or more) groups: one (the experimental group) receiving the treatment that is being tested, and the other (the comparison or control group) receiving an alternative treatment, a placebo (dummy treatment) or no treatment. The two groups are followed up to compare differences in outcomes to see how effective the experimental treatment was. Through randomisation, the groups should be similar in all aspects apart from the treatment they receive during the study.

Relative risk A summary measure which represents the ratio of the risk of a given event or outcome (for example, an adverse reaction to the drug being tested) in one group of subjects compared with another group. When the 'risk' of the event is the same in the two groups the relative risk is 1. In a study comparing two treatments, a relative risk of 2 would indicate that patients receiving one of the treatments had twice the risk of an undesirable outcome than those receiving the other treatment. Relative risk is sometimes used as a synonym for risk ratio.

Systematic review A review in which evidence from scientific studies has been identified, appraised and synthesised in a methodical way according to predetermined criteria. May or may not include a meta-analysis.

2307

2308 Please see the NICE glossary 2309 (www.nice.org.uk/website/glossary/glossary.jsp) for an explanation of terms 2310 not described above.

2311 Abbreviations

Abbreviation Term ARI Absolute risk increase Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 145 of 150

ARR Absolute risk reduction CI Confidence interval GRADE Grading of Recommendations Assessment, Development and Evaluation ICER Incremental cost-effectiveness ratio IMMPACT Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials NNTB Number needed to treat to benefit NNTH Number needed to treat to harm PDN Painful diabetic neuropaty PHN Post herpetic neuralgia QALY Quality-adjusted life year RCT Randomised controlled trial RR Relative risk SD Standard deviation SE Standard error 2312

2313

2314 Appendix A Contributors and declarations of interests

2315 The Guideline Development Group

2316 Peter Barry (Chair) 2317 Consultant in Paediatric Intensive Care, University Hospitals of Leicester NHS 2318 Trust and Honorary Senior Lecturer, Department of Child Health, University of 2319 Leicester

2320 Tracey Cole 2321 Patient and carer representative

2322 Paula Crawford 2323 Lead pharmacist therapeutic review, Belfast HSC Trust

2324 Peter Crome 2325 Professor of Geriatric Medicine, Keele University and Consultant Geriatrician, 2326 North Staffordshire Combined Healthcare NHS Trust

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 146 of 150

2327 Matthew Doyle 2328 GP Partner, Cromwell Place Surgery, St Ives

2329 Niru Goenka 2330 Consultant Physician, Countess of Chester NHS Foundation Trust

2331 Clair Haslam (resigned from GDG after meeting 3) 2332 Nurse Consultant, Pain Management and Neuromodulation, The Walton 2333 Centre, Liverpool

2334 John Lee 2335 Consultant in Pain Medicine, University College London (UCL) Hospitals and 2336 Honorary Senior Lecturer, UCL

2337 Chris McDermott 2338 Clinical Senior Lecturer and Honorary Consultant Neurologist, Royal 2339 Hallamshire Hospital

2340 Vera Neumann 2341 Consultant and Honorary Senior Lecturer in Rehabilitation Medicine, Leeds 2342 Teaching Hospitals NHS Trust and University of Leeds

2343 David Rowbotham 2344 Professor of Anaesthesia and Pain Management, Department of Health 2345 Sciences, Leicester Medical School, University of Leicester

2346 Blair H. Smith 2347 Professor of Population Science, University of Dundee, and GP, Peterhead 2348 Medical Practice

2349 Heather Wallace 2350 Patient and carer representative

2351 Co-opted member

2352 The following person was not a full member of the Guideline Development 2353 Group but was co-opted onto the group as an expert adviser:

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 147 of 150

2354 Soloman Tesfaye 2355 Consultant Physician/Endocrinologist, Royal Hallamshire Hospital, Sheffield

2356 Short Clinical Guidelines Technical Team

2357 A Short Clinical Guidelines Technical team was responsible for this guideline 2358 throughout its development. It prepared information for the Guideline 2359 Development Group, drafted the guideline and responded to consultation 2360 comments.

2361 Steven Barnes 2362 Technical Analyst

2363 Kathryn Chamberlain 2364 Project Manager

2365 Nicole Elliott 2366 Associate Director

2367 Sarah Glover 2368 Information Specialist

2369 Michael Heath 2370 Programme Manager

2371 Prashanth Kandaswamy 2372 Senior Technical Adviser – Health economics

2373 Victoria Kelly 2374 Project Manager

2375 Toni Tan 2376 Technical Advisor

2377 The Guideline Review Panel

2378 The Guideline Review Panel is an independent panel that oversees the 2379 development of the guideline and takes responsibility for monitoring 2380 adherence to NICE guideline development processes. In particular, the panel

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 148 of 150

2381 ensures that stakeholder comments have been adequately considered and 2382 responded to. The panel includes members from the following perspectives: 2383 primary care, secondary care, lay, public health and industry.

2384 Robert Walker (Chair) 2385 General Practitioner, Workington

2386 John Harley 2387 Clinical Governance and Prescribing Lead and General Practitioner, North 2388 Tees PCT

2389 Ailsa Donnelly 2390 Lay member

2391 Sarah Fishburn 2392 Lay member

2393 NICE Centre for Clinical Practice

2394 Fergus Macbeth 2395 Director, Centre for Clinical Practice

2396 Emma Banks 2397 Guidelines Coordinator

2398 Stefanie Reken (née Kuntze) 2399 Technical Analyst (Health Economics)

2400 Rachel Ryle 2401 Guidelines Commissioning Manager

2402 Beth Shaw 2403 Senior Technical Adviser

2404 Judith Thornton 2405 Technical Analyst

2406 Claire Turner 2407 Guidelines Commissioning Manager (from September 2009)

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 149 of 150

2408 Editors

2409 Lyn Knott 2410 Senior Medical Editor

2411 Lynne Kincaid 2412 Medical Editor

2413 Declarations of interests

2414 [Add declarations here]

2415

2416

Neuropathic pain: NICE clinical guideline DRAFT (September 2011) 150 of 150